Search

Simple Summary: The increased detection of pancreatic cysts in recent years has triggered extensive diagnostic investigations to clarify their potential risk of malignancy, resulting in a large number of patients undergoing numerous imaging follow-up studies for many years. Therefore, there is a growing need for optimization of the current surveillance protocol to provide a practical and reasonable solution in the face of an ever-growing demand. The aim of this paper is to discuss the current available evidence on whether the implementation of unenhanced abbreviated-MRI (A-MRI) protocols for cystic pancreatic lesion surveillance could improve healthcare economics and reduce waiting lists in clinical practice without significantly reducing diagnostic accuracy. The increased detection of pancreatic cysts in recent years has triggered extensive diagnostic investigations to clarify their potential risk of malignancy, resulting in a large number of patients undergoing numerous imaging follow-up studies for many years. Therefore, there is a growing need for optimization of the current surveillance protocol to reduce both healthcare costs and waiting lists, while still maintaining appropriate sensibility and specificity. Imaging is an essential tool for evaluating patients with intraductal papillary mucinous neoplasms (IPMNs) since it can assess several predictors for malignancy and thus guide further management recommendations. Although contrast-enhanced magnetic resonance imaging (MRI) with magnetic resonance cholangiopancreatography (MRCP) has been widely recommended by most international guidelines, recent results support the use of unenhanced abbreviated-MRI (A-MRI) protocols as a surveillance tool in patients with IPMN. In fact, A-MRI has shown high diagnostic performance in malignant detection, with high sensitivity and specificity as well as excellent interobserver agreement. The aim of this paper is, therefore, to discuss the current available evidence on whether the implementation of an abbreviated-MRI (A-MRI) protocol for cystic pancreatic lesion surveillance could improve healthcare economics and reduce waiting lists in clinical practice without significantly reducing diagnostic accuracy. [ABSTRACT FROM AUTHOR]

Simple Summary: This paper delves into the persistent issue of unequal access to medical imaging, with a particular focus on breast cancer screening and its impact on marginalized communities and racial/ethnic minorities. Central to our discussion is the role of scientific mobility among radiologists in fostering healthcare policy changes that promote diversity and cultural competence. We propose various strategies to bridge this gap, including cultural education, sensitivity training, and diversifying the radiology workforce. These measures aim to improve communication with diverse patient groups and reduce healthcare disparities. Additionally, we explore the challenges and advantages of teleradiology as a means to extend medical imaging services to underserved areas. In the context of artificial intelligence, we emphasize the critical need to validate algorithms across diverse populations to ensure unbiased and equitable healthcare outcomes. Overall, this paper underscores the importance of international collaboration in addressing global access barriers, presenting it as a key to mitigating disparities in medical imaging access and contributing to the pursuit of equitable healthcare. Access to medical imaging is pivotal in healthcare, playing a crucial role in the prevention, diagnosis, and management of diseases. However, disparities persist in this scenario, disproportionately affecting marginalized communities, racial and ethnic minorities, and individuals facing linguistic or cultural barriers. This paper critically assesses methods to mitigate these disparities, with a focus on breast cancer screening. We underscore scientific mobility as a vital tool for radiologists to advocate for healthcare policy changes: it not only enhances diversity and cultural competence within the radiology community but also fosters international cooperation and knowledge exchange among healthcare institutions. Efforts to ensure cultural competency among radiologists are discussed, including ongoing cultural education, sensitivity training, and workforce diversification. These initiatives are key to improving patient communication and reducing healthcare disparities. This paper also highlights the crucial role of policy changes and legislation in promoting equal access to essential screening services like mammography. We explore the challenges and potential of teleradiology in improving access to medical imaging in remote and underserved areas. In the era of artificial intelligence, this paper emphasizes the necessity of validating its models across a spectrum of populations to prevent bias and achieve equitable healthcare outcomes. Finally, the importance of international collaboration is illustrated, showcasing its role in sharing insights and strategies to overcome global access barriers in medical imaging. Overall, this paper offers a comprehensive overview of the challenges related to disparities in medical imaging access and proposes actionable strategies to address these challenges, aiming for equitable healthcare delivery. [ABSTRACT FROM AUTHOR]

Simple Summary: Skin diseases can be serious, and early detection is key to effective treatment. Unfortunately, the quality of images used to diagnose these diseases often suffers due to interference from hair, making accurate diagnosis challenging. This research introduces a novel technology, the DM–AHR, a self-supervised conditional diffusion model designed specifically to generate clear, hairless images for better skin disease diagnosis. Our work not only presents a new, advanced model that expertly identifies and removes hair from dermoscopic images but also introduces a specialized dataset, DERMAHAIR, to further research and improve diagnostic processes. The enhancements in image quality provided by DM–AHR significantly improve the accuracy of skin disease diagnoses, and it promises to be a valuable tool in medical imaging. Accurate skin diagnosis through end-user applications is important for early detection and cure of severe skin diseases. However, the low quality of dermoscopic images hampers this mission, especially with the presence of hair on these kinds of images. This paper introduces DM–AHR, a novel, self-supervised conditional diffusion model designed specifically for the automatic generation of hairless dermoscopic images to improve the quality of skin diagnosis applications. The current research contributes in three significant ways to the field of dermatologic imaging. First, we develop a customized diffusion model that adeptly differentiates between hair and skin features. Second, we pioneer a novel self-supervised learning strategy that is specifically tailored to optimize performance for hairless imaging. Third, we introduce a new dataset, named DERMAHAIR (DERMatologic Automatic HAIR Removal Dataset), that is designed to advance and benchmark research in this specialized domain. These contributions significantly enhance the clarity of dermoscopic images, improving the accuracy of skin diagnosis procedures. We elaborate on the architecture of DM–AHR and demonstrate its effective performance in removing hair while preserving critical details of skin lesions. Our results show an enhancement in the accuracy of skin lesion analysis when compared to existing techniques. Given its robust performance, DM–AHR holds considerable promise for broader application in medical image enhancement. [ABSTRACT FROM AUTHOR]

Simple Summary: HER2-positive breast cancer occurs in 15–30% of cases and has a poor prognosis. Digital image analysis of HER2 is promising, but its implementation in real clinical practice remains unclear. This systematic review evaluates the effectiveness of digital image analysis algorithms for HER2 in breast cancer and their performance, with a focus on testing them in real-world clinical settings. The authors aim to assess the applicability of these algorithms in practical clinical scenarios. By analyzing 25 papers from the period 2013–2024 and emphasizing mostly deep learning approaches, the review underscores the importance of standardized evaluation criteria, study designs tailored for clinical applications, and clinical validation. While direct evidence of clinical application was not found, the findings aim to guide future research and the implementation of digital image analysis in breast cancer diagnosis within clinical settings, potentially impacting the research community by advancing algorithmic applications in real clinical practice. This systematic review aims to address the research gap in the performance of computational algorithms for the digital image analysis of HER2 images in clinical settings. While numerous studies have explored various aspects of these algorithms, there is a lack of comprehensive evaluation regarding their effectiveness in real-world clinical applications. We conducted a search of the Web of Science and PubMed databases for studies published from 31 December 2013 to 30 June 2024, focusing on performance effectiveness and components such as dataset size, diversity and source, ground truth, annotation, and validation methods. The study was registered with PROSPERO (CRD42024525404). Key questions guiding this review include the following: How effective are current computational algorithms at detecting HER2 status in digital images? What are the common validation methods and dataset characteristics used in these studies? Is there standardization of algorithm evaluations of clinical applications that can improve the clinical utility and reliability of computational tools for HER2 detection in digital image analysis? We identified 6833 publications, with 25 meeting the inclusion criteria. The accuracy rate with clinical datasets varied from 84.19% to 97.9%. The highest accuracy was achieved on the publicly available Warwick dataset at 98.8% in synthesized datasets. Only 12% of studies used separate datasets for external validation; 64% of studies used a combination of accuracy, precision, recall, and F1 as a set of performance measures. Despite the high accuracy rates reported in these studies, there is a notable absence of direct evidence supporting their clinical application. To facilitate the integration of these technologies into clinical practice, there is an urgent need to address real-world challenges and overreliance on internal validation. Standardizing study designs on real clinical datasets can enhance the reliability and clinical applicability of computational algorithms in improving the detection of HER2 cancer. [ABSTRACT FROM AUTHOR]

Simple Summary: Breast cancer is a significant cause of female cancer-related deaths in the US. Checking how severe the cancer is helps in planning treatment. Modern AI methods are good at grading cancer, but they are not used much in hospitals yet. We developed and utilized ensemble deep learning algorithms for addressing the tasks of classifying (1) breast cancer subtype and (2) breast cancer invasiveness from whole slide image (WSI) histopathology slides. The ensemble models used were based on convolutional neural networks (CNNs) known for extracting distinctive features crucial for accurate classification. In this paper, we provide a comprehensive analysis of these models and the used methodology for breast cancer diagnosis tasks. Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, a comprehensive approach is presented utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets were based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH dataset, a proposed ensemble strategy was employed, incorporating VGG16 and ResNet50 architectures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique to preprocess a high-resolution image facilitated a focused analysis of localized regions of interest. The annotated BACH dataset encompassed 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, the proposed ensemble was used on the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify microscopic images into eight distinct categories (four benign and four malignant). For both datasets, a five-fold cross-validation approach was employed for rigorous training and testing. Preliminary experimental results indicated a patch classification accuracy of 95.31% (for the BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to ongoing endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens. [ABSTRACT FROM AUTHOR]

Simple Summary: Pencil beam scanning nozzles monitor the beam position in real time and record the results in log files. We cannot, however, place a beam position monitor at the isocenter during treatment, so accurate online beam position corrections and log file analyses rely on an algorithm to project the beam position from the nozzle to the isocentric plane. We present four generic algorithms and determined the accuracy of each approach in three example configurations and two nozzle lengths. Beam position uncertainties along the beam trajectory arise from the accelerator, beamline, and scanning magnets (SMs). They can be monitored in real time, e.g., through strip ionization chambers (ICs), and treatments can be paused if needed. Delivery is more reliable and accurate if the beam position is projected from monitored nozzle parameters to the isocenter, allowing for accurate online corrections to be performed. Beam position projection algorithms are also used in post-delivery log file analyses. In this paper, we investigate the four potential algorithms that can be applied to all pencil beam scanning (PBS) nozzles. For some combinations of nozzle configurations and algorithms, however, the projection uses beam properties determined offline (e.g., through beam tuning or technical commissioning). The best algorithm minimizes either the total uncertainty (i.e., offline and online) or the total offline uncertainty in the projection. Four beam position algorithms are analyzed (A1–A4). Two nozzle lengths are used as examples: a large nozzle (1.5 m length) and a small nozzle (0.4 m length). Three nozzle configurations are considered: IC after SM, IC before SM, and ICs on both sides. Default uncertainties are selected for ion chamber measurements, nozzle entrance beam position and angle, and scanning magnet angle. The results for other uncertainties can be determined by scaling these results or repeating the error propagation. We show the propagation of errors from two locations and the SM angle to the isocenter for all the algorithms. The best choice of algorithm depends on the nozzle length and is A1 and A3 for the large and small nozzles, respectively. If the total offline uncertainty is to be minimized (a better choice if the offline uncertainty is not stable), the best choice of algorithm changes to A1 for the small nozzle for some hardware configurations. Reducing the nozzle length can help to reduce the gantry size and make proton therapy more accessible. This work is important for designing smaller nozzles and, consequently, smaller gantries. This work is also important for log file analyses. [ABSTRACT FROM AUTHOR]

Simple Summary: Risk assessment in early breast cancer is critical for clinical decisions, but defining risk categories poses a significant challenge. The integration of conventional histopathology and biomarkers with artificial intelligence (AI) techniques, including machine learning and deep learning, has the potential to offer more precise information. AI applications extend beyond detection to histological subtyping, grading, and molecular feature identification. The successful integration of AI into clinical practice requires collaboration between histopathologists, molecular pathologists, computational pathologists, and oncologists to optimize patient outcomes. Effective risk assessment in early breast cancer is essential for informed clinical decision-making, yet consensus on defining risk categories remains challenging. This paper explores evolving approaches in risk stratification, encompassing histopathological, immunohistochemical, and molecular biomarkers alongside cutting-edge artificial intelligence (AI) techniques. Leveraging machine learning, deep learning, and convolutional neural networks, AI is reshaping predictive algorithms for recurrence risk, thereby revolutionizing diagnostic accuracy and treatment planning. Beyond detection, AI applications extend to histological subtyping, grading, lymph node assessment, and molecular feature identification, fostering personalized therapy decisions. With rising cancer rates, it is crucial to implement AI to accelerate breakthroughs in clinical practice, benefiting both patients and healthcare providers. However, it is important to recognize that while AI offers powerful automation and analysis tools, it lacks the nuanced understanding, clinical context, and ethical considerations inherent to human pathologists in patient care. Hence, the successful integration of AI into clinical practice demands collaborative efforts between medical experts and computational pathologists to optimize patient outcomes. [ABSTRACT FROM AUTHOR]

Simple Summary: A large body of multidisciplinary evidence involves the topic of thyroid cancer (the most common endocrine malignancy). Nevertheless, exceptional findings such as thyroid cancer in ectopic thyroid tissue, representing 0.3–0.5% of the malignant neoplasia with any location, suggest even greater challenges. Awareness remains the key operative element since the index of suspicion is low, especially in non-cervical areas. Hence, currently, the ectopic thyroid remains a matter of individualized management. The ectopic mediastinal thyroid (EMT) is part of the less frequent sublingual ectopic sites. Here, we introduce the most complex analysis in published EMT data (N = 117 patients) that identified an unexpectedly high rate of malignancy (18.8%), papillary cancer being the most frequent histological type. A rate of 5.98% amid all EMTs represented individuals confirmed with unrelated (non-thyroid) malignancies. Thyroid anomalies (other than EMT presence) were reported in 38.33% of the benign EMT, while the overall malignancy rate in EMTs was higher than expected according to prior data when compared to other ectopic sites. We aimed to analyze the management of the ectopic mediastinal thyroid (EMT) with respect to EMT-related cancer and non-malignant findings related to the pathological report, clinical presentation, imaging traits, endocrine profile, connective tissue to the cervical (eutopic) thyroid gland, biopsy or fine needle aspiration (FNA) results, surgical techniques and post-operatory outcome. This was a comprehensive review based on revising any type of freely PubMed-accessible English, full-length original papers including the keywords "ectopic thyroid" and "mediastinum" from inception until March 2024. We included 89 original articles that specified EMTs data. We classified them into four main groups: (I) studies/case series (n = 10; N = 36 EMT patients); (II) malignant EMTs (N = 22 subjects; except for one newborn with immature teratoma in the EMT, only adults were reported; mean age of 62.94 years; ranges: 34 to 90 years; female to male ratio of 0.9). Histological analysis in adults showed the following: papillary (N = 11/21); follicular variant of the papillary type (N = 2/21); Hürthle cell thyroid follicular malignancy (N = 1/21); poorly differentiated (N = 1/21); anaplastic (N = 2/21); medullary (N = 1/21); lymphoma (N = 2/21); and MALT (mucosa-associated lymphoid tissue) (N = 1/21); (III) benign EMTs with no thyroid anomalies (N = 37 subjects; mean age of 56.32 years; ranges: 30 to 80 years; female to male ratio of 1.8); (IV) benign EMTs with thyroid anomalies (N = 23; female to male ratio of 5.6; average age of 52.1 years). This panel involved clinical/subclinical hypothyroidism (iatrogenic, congenital, thyroiditis-induced, and transitory type upon EMT removal); thyrotoxicosis (including autonomous activity in EMTs that suppressed eutopic gland); autoimmune thyroiditis/Graves's disease; nodules/multinodular goiter and cancer in eutopic thyroid or prior thyroidectomy (before EMT detection). We propose a 10-item algorithm that might help navigate through the EMT domain. To conclude, across this focused-sample analysis (to our knowledge, the largest of its kind) of EMTs, the EMT clinical index of suspicion remains low; a higher rate of cancer is reported than prior data (18.8%), incident imagery-based detection was found in 10–14% of the EMTs; surgery offered an overall good outcome. A wide range of imagery, biopsy/FNA and surgical procedures is part of an otherwise complex personalized management. [ABSTRACT FROM AUTHOR]

Simple Summary: The 2021 WHO classification of central nervous system (CNS) tumors is challenging for neuroradiologists due to the central role of the molecular profile of tumors. We performed a scoping review of recent literature to assess the existing data on the power of novel data analysis tools to predict new tumor classes by imaging. We found room for performance improvement for subgroups with lower incidence (e.g., 1p/19q codeleted or IDH1/2 mutated gliomas) and patients with rare diagnoses (e.g., pediatric gliomas, midline gliomas). More data regarding functional MRI techniques need to be collected. Studies explicitly designed to assess the generalizability of AI-aided tools for predicting molecular tumor subgroups are lacking. The 2021 WHO classification of CNS tumors is a challenge for neuroradiologists due to the central role of the molecular profile of tumors. The potential of novel data analysis tools in neuroimaging must be harnessed to maintain its role in predicting tumor subgroups. We performed a scoping review to determine current evidence and research gaps. A comprehensive literature search was conducted regarding glioma subgroups according to the 2021 WHO classification and the use of MRI, radiomics, machine learning, and deep learning algorithms. Sixty-two original articles were included and analyzed by extracting data on the study design and results. Only 8% of the studies included pediatric patients. Low-grade gliomas and diffuse midline gliomas were represented in one-third of the research papers. Public datasets were utilized in 22% of the studies. Conventional imaging sequences prevailed; data on functional MRI (DWI, PWI, CEST, etc.) are underrepresented. Multiparametric MRI yielded the best prediction results. IDH mutation and 1p/19q codeletion status prediction remain in focus with limited data on other molecular subgroups. Reported AUC values range from 0.6 to 0.98. Studies designed to assess generalizability are scarce. Performance is worse for smaller subgroups (e.g., 1p/19q codeleted or IDH1/2 mutated gliomas). More high-quality study designs with diversity in the analyzed population and techniques are needed. [ABSTRACT FROM AUTHOR]

Simple Summary: A novel approach for automatic detection of neoplastic lesions in lymph nodes is presented, which incorporates machine learning methods and the new LN-RADS scale. The presented solution incorporates different network structures with diverse datasets to improve the overall effectiveness. Final findings demonstrate that incorporating the LN-RADS scale labels improved the overall diagnosis, especially when compared with current, standard practices. The presented solution is meant as an aid in the diagnosis process. The paper presents a novel approach for the automatic detection of neoplastic lesions in lymph nodes (LNs). It leverages the latest advances in machine learning (ML) with the LN Reporting and Data System (LN-RADS) scale. By integrating diverse datasets and network structures, the research investigates the effectiveness of ML algorithms in improving diagnostic accuracy and automation potential. Both Multinominal Logistic Regression (MLR)-integrated and fully connected neuron layers are included in the analysis. The methods were trained using three variants of combinations of histopathological data and LN-RADS scale labels to assess their utility. The findings demonstrate that the LN-RADS scale improves prediction accuracy. MLR integration is shown to achieve higher accuracy, while the fully connected neuron approach excels in AUC performance. All of the above suggests a possibility for significant improvement in the early detection and prognosis of cancer using AI techniques. The study underlines the importance of further exploration into combined datasets and network architectures, which could potentially lead to even greater improvements in the diagnostic process. [ABSTRACT FROM AUTHOR]

Simple Summary: Small bowel tumors (SBTs) are rare entities with a steadily rising incidence in the last decades. The introduction of small bowel capsule endoscopy (SBCE) at the dawn of the new millennium represented a milestone in the early diagnosis and management of SBTs: this game-changing tool allowed clinicians to explore and manage pathology in previously inaccessible tracts of the digestive system. SBTs, encompassing various histological types, pose diagnostic challenges due to their rarity and non-specific clinical presentation. Early detection, crucial for prognosis, often requires a combination of diagnostic techniques, including cross-sectional imaging. With this narrative review, we aim to provide an updated insight into the significance of SBCE in addressing the complex management of SBTs. Small bowel tumors (SBT) are relatively rare, but have had a steadily increasing incidence in the last few decades. Small bowel capsule endoscopy (SBCE) and device-assisted enteroscopy are the main endoscopic techniques for the study of the small bowel, the latter additionally providing sampling and therapeutic options, and hence acting complementary to SBCE in the diagnostic work-up. Although a single diagnostic modality is often insufficient in the setting of SBTs, SBCE is a fundamental tool to drive further management towards a definitive diagnosis. The aim of this paper is to provide a concise narrative review of the role of SBCE in the diagnosis and management of SBTs. [ABSTRACT FROM AUTHOR]

Simple Summary: Personalised cancer treatment improves outcome for patients but requires that each individual tumour is classified using molecular tools. Extensive genomic studies for each individual patients are out of reach for the majority of the population, therefore, pathologists need to apply readily available tests to achieve molecular classification for all cancer patients. This paper describes how gastric cancer can be classified using on slide tests already used by histopathologists. Using this classification, oncologists can select more effective treatment for each patient. Background and Objectives: Gastric cancer (GC) is one of the most commonly diagnosed cancers and the fourth cause of cancer death worldwide. Personalised treatment improves GC outcomes. A molecular classification is needed to choose the appropriate therapy. A classification that uses on-slide biomarkers and formalin-fixed and paraffin-embedded (FFPE) tissue is preferable to comprehensive genomic analysis. In 2016, Setia and colleagues proposed an on-slide classification; however, this is not in widespread use. We propose a modification of this classification that has six subgroups: GC associated with Epstein–Barr virus (GC EBV+), GC with mismatch-repair deficiency (GC dMMR), GC with epithelial–mesenchymal transformation (GC EMT), GC with chromosomal instability (GC CIN), CG that is genomically stable (GC GS) and GC not otherwise specified (GC NOS). This classification also has a provision for biomarkers for current or emerging targeted therapies (Her2, PD-L1 and Claudin18.2). Here, we assess the implementation and feasibility of this inclusive working classification. Materials and Methods: We constructed a tissue microarray library from a cohort of 79 resection cases from FFPE tissue archives. We used a restricted panel of on-slide markers (EBER, MMR, E-cadherin, beta-catenin and p53), defined their interpretation algorithms and assigned each case to a specific molecular subtype. Results: GC EBV(+) cases were 6%, GC dMMR cases were 20%, GC EMT cases were 14%, GC CIN cases were 23%, GC GS cases were 29%, and GC NOS cases were 8%. Conclusions: This working classification uses markers that are widely available in histopathology and are easy to interpret. A diagnostic subgroup is obtained for 92% of the cases. The proportion of cases in each subgroup is in keeping with other published series. Widescale implementation appears feasible. A study using endoscopic biopsies is warranted. [ABSTRACT FROM AUTHOR]

Simple Summary: Early detection of melanoma and differentiation from benign nevi can be challenging even for the most experienced dermatologists. To improve melanoma detection, artificial intelligence algorithms incorporating dermoscopy have been developed, but lack transparency and therefore have limited training value for healthcare providers. To address this, an automated approach utilizing imaging biomarker cues (IBCs), logical features extracted from images that mimic expert dermatologists' dermoscopic pattern recognition skills, was developed. This study excluded deep learning approaches to which IBCs are complementary or alternative. Ten participants assessed 78 dermoscopic images (39 melanomas and 39 nevi) first without IBCs and then with IBCs. Using IBCs significantly improved diagnostic accuracy: sensitivity increased significantly from 73.69% to 81.57% (p = 0.0051) and specificity increased from 60.50% to 67.25% (p = 0.059). These results indicate that incorporating IBCs can significantly enhance melanoma diagnosis, with potential implications for improved screening practices. Further research is needed to confirm these findings across a variety of healthcare providers. Incorporation of dermoscopy and artificial intelligence (AI) is improving healthcare professionals' ability to diagnose melanoma earlier, but these algorithms often suffer from a "black box" issue, where decision-making processes are not transparent, limiting their utility for training healthcare providers. To address this, an automated approach for generating melanoma imaging biomarker cues (IBCs), which mimics the screening cues used by expert dermoscopists, was developed. This study created a one-minute learning environment where dermatologists adopted a sensory cue integration algorithm to combine a single IBC with a risk score built on many IBCs, then immediately tested their performance in differentiating melanoma from benign nevi. Ten participants evaluated 78 dermoscopic images, comprised of 39 melanomas and 39 nevi, first without IBCs and then with IBCs. Participants classified each image as melanoma or nevus in both experimental conditions, enabling direct comparative analysis through paired data. With IBCs, average sensitivity improved significantly from 73.69% to 81.57% (p = 0.0051), and the average specificity improved from 60.50% to 67.25% (p = 0.059) for the diagnosis of melanoma. The index of discriminability (d′) increased significantly by 0.47 (p = 0.002). Therefore, the incorporation of IBCs can significantly improve physicians' sensitivity in melanoma diagnosis. While more research is needed to validate this approach across other healthcare providers, its use may positively impact melanoma screening practices. [ABSTRACT FROM AUTHOR]

Simple Summary: Cancer develops through a complex process involving genetic changes that can lead to uncontrolled cell growth and tumor formation. This research focuses on developing an advanced approach to classify tumors into meaningful subgroups based on somatic mutations. Using machine learning techniques, specifically a deep neural network, and integrating genetic data with known gene interaction networks, we propose a framework for tumor stratification, called D3NS (deep neural network integrated into network-based stratification). This framework identifies patient subtypes predictive for survival and significantly associated with several clinical outcomes (tumor stage, grade and treatment response). We applied D3NS to real-world data from the Cancer Genome Atlas for bladder, ovarian, and kidney cancers. The results demonstrate the potential of this approach to improve cancer stratification, positioning it as a useful base model for cancer research and a promising tool in clinical settings. (1) Background: The identification of tumor subtypes is fundamental in precision medicine for accurate diagnoses and personalized therapies. Cancer development is often driven by the accumulation of somatic mutations that can cause alterations in tissue functions and morphologies. In this work, a method based on a deep neural network integrated into a network-based stratification framework (D3NS) is proposed to stratify tumors according to somatic mutations. (2) Methods: This approach leverages the power of deep neural networks to detect hidden information in the data by combining the knowledge contained in a network of gene interactions, as typical of network-based stratification methods. D3NS was applied using real-world data from The Cancer Genome Atlas for bladder, ovarian, and kidney cancers. (3) Results: This technique allows for the identification of tumor subtypes characterized by different survival rates and significant associations with several clinical outcomes (tumor stage, grade or response to therapy). (4) Conclusion: D3NS can provide a base model in cancer research and could be considered as a useful tool for tumor stratification, offering potential support in clinical settings. [ABSTRACT FROM AUTHOR]

Simple Summary: Epithelioid vascular tumors are rare bone vascular lesions with a spectrum of overlapping cancer entities with varying degrees of malignant potential that remain controversial because of their rarity, unusual morphologic features, and unpredictable biologic behavior. The classification of vascular tumors of bone proposed by the World Health Organization in 2020 incorporated morphologic findings with available genetic evidence, including mainly angiosarcoma, epithelioid hemangioma, pseudomyogenic hemangioendothelioma, and epithelioid hemangioendothelioma. Here, we report our experience with eleven patients suffering from epithelioid hemangioma (EH) of the spine. As very few cases of EH of the spine have been reported in the literature, the evidence-based decision-making process for these patients can be very difficult. Thus, we propose a treatment flowchart based on our single center's experience. Epithelioid hemangioma is recognized by the World Health Organization as a distinct benign neoplasm; however, it is characterized by locally aggressive and rarely metastasizing behavior. Epithelioid vascular tumors are rare bony vascular lesions with varying degrees of malignant potential that remain controversial because of their rarity, unusual morphological features, and unpredictable biological behavior. The application of new molecular tools, such as massive parallel sequencing technologies, have provided new diagnostic markers and an opportunity to further refine the classification of bone vascular neoplasms. Very few cases of EH of the spine have been reported in the literature; therefore, it is difficult to make evidence-based therapeutic decisions for these patients. We report herein our experience with eleven patients suffering from EH of the spine. The study population included three males and eight females treated in our center from 2016 to the present; the average age was 44.8 years (range 14–75 years). The surgical, clinical, and radiographic data were retrospectively analyzed. The mean follow-up was 34.8 months. All patients presented lytic vertebral body lesions, six of them with pathological fracture. The majority of patients (80%) presented myelo-radicular compression. All patients were surgically treated, and preoperative embolization was performed in all cases. In light of the literature review and the clinical experience of our center, we can consider EH a locally aggressive tumor that requires surgical treatment in case of symptoms. Here, we propose a treatment algorithm that could be useful in the management of patients with this rare disease. [ABSTRACT FROM AUTHOR]

Simple Summary: Pancreatic neuroendocrine tumors are rare, but their incidence is rising. Several grades exist, and distinguishing between these is pivotal for clinical management. Currently, the grades can only be differentiated by histological analysis requiring invasive sampling. MicroRNAs are short non-protein coding RNA molecules that were shown to be differentially expressed in a wide variety of tumors. Here, we examined whether microRNAs could be exploited to differentiate grade 1 and 2 pancreatic neuroendocrine tumors and established significantly differentially expressed microRNAs. Pancreatic neuroendocrine neoplasms pose a growing clinical challenge due to their rising incidence and variable prognosis. The current study aims to investigate microRNAs (miRNA; miR) as potential biomarkers for distinguishing between grade 1 (G1) and grade 2 (G2) pancreatic neuroendocrine tumors (PanNET). A total of 33 formalin-fixed, paraffin-embedded samples were analyzed, comprising 17 G1 and 16 G2 tumors. Initially, literature-based miRNAs were validated via real-time quantitative reverse transcription polymerase chain reaction (RT-qPCR), confirming significant downregulation of miR-130b-3p and miR-106b in G2 samples. Through next-generation sequencing, we have identified and selected the top six miRNAs showing the highest difference between G1 and G2 tumors, which were further validated. RT-qPCR validation confirmed the downregulation of miR-30d-5p in G2 tumors. miRNA combinations were created to distinguish between the two PanNET grades. The highest diagnostic performance in distinguishing between G1 and G2 PanNETs by a machine learning algorithm was achieved when using the combination miR-106b + miR-130b-3p + miR-127-3p + miR-129-5p + miR-30d-5p. The ROC analysis resulted in a sensitivity of 83.33% and a specificity of 87.5%. The findings underscore the potential use of miRNAs as biomarkers for stratifying PanNET grades, though further research is warranted to enhance diagnostic accuracy and clinical utility. [ABSTRACT FROM AUTHOR]

Simple Summary: A 90-day mortality predictive model for curative gastric cancer resection based on the Spanish EURECCA Esophagogastric Cancer database was externally validated using the GASTRODATA registry. The externally validated model showed a modestly worse performance compared to the original model, nevertheless maintaining its discriminating ability in clinical practice. Background: Radical gastrectomy remains the main treatment for gastric cancer, despite its high mortality. A clinical predictive model of 90-day mortality (90DM) risk after gastric cancer surgery based on the Spanish EURECCA registry database was developed using a matching learning algorithm. We performed an external validation of this model based on data from an international multicenter cohort of patients. Methods: A cohort of patients from the European GASTRODATA database was selected. Demographic, clinical, and treatment variables in the original and validation cohorts were compared. The performance of the model was evaluated using the area under the curve (AUC) for a random forest model. Results: The validation cohort included 2546 patients from 24 European hospitals. The advanced clinical T- and N-category, neoadjuvant therapy, open procedures, total gastrectomy rates, and mean volume of the centers were significantly higher in the validation cohort. The 90DM rate was also higher in the validation cohort (5.6%) vs. the original cohort (3.7%). The AUC in the validation model was 0.716. Conclusion: The externally validated model for predicting the 90DM risk in gastric cancer patients undergoing gastrectomy with curative intent continues to be as useful as the original model in clinical practice. [ABSTRACT FROM AUTHOR]

Simple Summary: This research aims to evaluate the effectiveness of a machine learning algorithm, XGBoost, in predicting treatment recommendations for patients with hepatocellular carcinoma (HCC). The study uses clinical and imaging data from patients discussed at a multidisciplinary tumor board. The findings suggest that the algorithm can accurately predict all eight treatment recommendations made by the board, potentially aiding clinical decision-making in settings lacking subspecialty expertise. This study addresses the potential of machine learning in predicting treatment recommendations for patients with hepatocellular carcinoma (HCC). Using an IRB-approved retrospective study of patients discussed at a multidisciplinary tumor board, clinical and imaging variables were extracted and used in a gradient-boosting machine learning algorithm, XGBoost. The algorithm's performance was assessed using confusion matrix metrics and the area under the Receiver Operating Characteristics (ROC) curve. The study included 140 patients (mean age 67.7 ± 8.9 years), and the algorithm was found to be predictive of all eight treatment recommendations made by the board. The model's predictions were more accurate than those based on published therapeutic guidelines by ESMO and NCCN. The study concludes that a machine learning model incorporating clinical and imaging variables can predict treatment recommendations made by an expert multidisciplinary tumor board, potentially aiding clinical decision-making in settings lacking subspecialty expertise. [ABSTRACT FROM AUTHOR]

Simple Summary: The incidence of renal cell carcinoma (RCC) is increasing due to the expansion of cross-sectional imaging and advanced imaging techniques. They allow for the detection of tumors at an earlier stage, but there are often overlapping similarities in the appearance of benign and malignant renal tumors. This review presents and discusses the ever-evolving landscape of imaging techniques that can be used to detect and diagnose renal cell carcinoma, including its major histologic subtypes. It also provides insight into recently proposed or updated imaging algorithms and guidelines for the diagnosis of RCC. The review considers the major advances in spectral CT, photo- counting CT, multiparametric MRI, contrast-enhanced ultrasound, sestamibi SPECT/CT, PSMA PET/CT, radiomics, artificial intelligence, Bosniak classification version 2019, clear cell likelihood score, and AUA guidelines. The goal for radiologists is to be better equipped to guide the diagnosis and management of these patients. This review highlights recent advances in renal cell carcinoma (RCC) imaging. It begins with dual-energy computed tomography (DECT), which has demonstrated a high diagnostic accuracy in the evaluation of renal masses. Several studies have suggested the potential benefits of iodine quantification, particularly for distinguishing low-attenuation, true enhancing solid masses from hyperdense cysts. By determining whether or not a renal mass is present, DECT could avoid the need for additional imaging studies, thereby reducing healthcare costs. DECT can also provide virtual unenhanced images, helping to reduce radiation exposure. The review then provides an update focusing on the advantages of multiparametric magnetic resonance (MR) imaging performance in the histological subtyping of RCC and in the differentiation of benign from malignant renal masses. A proposed standardized stepwise reading of images helps to identify clear cell RCC and papillary RCC with a high accuracy. Contrast-enhanced ultrasound may represent a promising diagnostic tool for the characterization of solid and cystic renal masses. Several combined pharmaceutical imaging strategies using both sestamibi and PSMA offer new opportunities in the diagnosis and staging of RCC, but their role in risk stratification needs to be evaluated. Although radiomics and tumor texture analysis are hampered by poor reproducibility and need standardization, they show promise in identifying new biomarkers for predicting tumor histology, clinical outcomes, overall survival, and the response to therapy. They have a wide range of potential applications but are still in the research phase. Artificial intelligence (AI) has shown encouraging results in tumor classification, grade, and prognosis. It is expected to play an important role in assessing the treatment response and advancing personalized medicine. The review then focuses on recently updated algorithms and guidelines. The Bosniak classification version 2019 incorporates MRI, precisely defines previously vague imaging terms, and allows a greater proportion of masses to be placed in lower-risk classes. Recent studies have reported an improved specificity of the higher-risk categories and better inter-reader agreement. The clear cell likelihood score, which adds standardization to the characterization of solid renal masses on MRI, has been validated in recent studies with high interobserver agreement. Finally, the review discusses the key imaging implications of the 2017 AUA guidelines for renal masses and localized renal cancer. [ABSTRACT FROM AUTHOR]

Simple Summary: Cancer genome analysis often reveals structural variants (SVs) involving fusion genes that are difficult to classify as drivers or passengers. Obtaining accurate AI predictions and explanations, which are crucial for a reliable diagnosis, is challenging. We developed an explainable AI (XAI) system that predicts the pathogenicity of SVs with gene fusions, providing reasons for its predictions. Our XAI achieved high accuracy, comparable to existing tools, and generated plausible explanations based on pathogenic mechanisms. This research represents a promising step towards AI-supported decision making in genomic medicine, enabling efficient and accurate diagnosis. When analyzing cancer sample genomes in clinical practice, many structural variants (SVs), other than single nucleotide variants (SNVs), have been identified. To identify driver variants, the leading candidates must be narrowed down. When fusion genes are involved, selection is particularly difficult, and highly accurate predictions from AI is important. Furthermore, we also wanted to determine how the prediction can make more reliable diagnoses. Here, we developed an explainable AI (XAI) suitable for SVs with gene fusions, based on the XAI technology we previously developed for the prediction of SNV pathogenicity. To cope with gene fusion variants, we added new data to the previous knowledge graph for SVs and we improved the algorithm. Its prediction accuracy was as high as that of existing tools. Moreover, our XAI could explain the reasons for these predictions. We used some variant examples to demonstrate that the reasons are plausible in terms of pathogenic basic mechanisms. These results can be seen as a hopeful step toward the future of genomic medicine, where efficient and correct decisions can be made with the support of AI. [ABSTRACT FROM AUTHOR]

Simple Summary: A new planning and optimization algorithm, VOLOTM Ultra, has been introduced for tomotherapy planning. This algorithm includes new features that significantly changed how treatments are planned and optimized. We have reoptimized 1056 treatment plans for 36 patients and six anatomical locations (breast with and without nodes, head and neck, cervix, esophagus, and hypofractionated). The results allowed us to characterize the impact of the accelerated treatment, leaf open/close time cutoff, normal tissue objective weight, number of optimization rounds, and use of rings on dosimetric quality and beam on time. The aim was to give a better understanding of the impact of the planning parameters on tomotherapy treatments to improve the quality of treatment for patients undergoing tomotherapy. Background: This work aimed to determine the optimum VOLOTM Ultra algorithm parameters for tomotherapy treatments. Methods: 1056 treatment plans were generated with VOLOTM Ultra for 36 patients and six anatomical locations. The impact of varying four parameters was studied: the accelerated treatment (AT), leaf open/close time (LOT) cutoff, normal tissue objective (NTO) weight, and number of iterations. The beam-on time and dosimetric metrics were quantified for the target volumes and organs at risk (OARs). Delivery quality assurance measurements were obtained for 36 plans to assess the delivery accuracy. Results: The mean beam-on time for the helical tomotherapy and TomoDirect (TD) plans decreased by 26.6 ± 2.8% and 17.4 ± 4.3%, respectively, when the accelerated treatment parameter was increased from 0 to 10, at the expense of the planning target volume (PTV) coverage (2% lower D98%) and OAR dose (up to 15% increase). For TD plans, it seems preferable to systematically use an AT value of 10. Increasing the number of iterations beyond six seems unnecessary. In this study, an NTO weight of approximately 10 appears to be ideal and eliminates the need to use rings in the treatment plan. Finally, no correlation was found between the leaf open/close time cutoff and the delivery accuracy, while a leaf open/close cutoff of 60 ms seemed to degrade dosimetry quality. Conclusion: Optimal values for the AT, LOT cutoff, NTO weight, and number of optimization rounds were identified and should help improve the management of patients whose tomotherapy treatments are planned with VOLOTM Ultra. [ABSTRACT FROM AUTHOR]

Simple Summary: Clear cell renal cell carcinoma (ccRCC) accounts for at least 80% of renal tumours worldwide. The grading of clear cell carcinoma is crucial for its management; therefore, it is important to distinguish the ccRCC grade pre-operatively. The aim of this research is to differentiate high- from low-grade ccRCC non-invasively using machine learning (ML) and radiomics features extracted from pre-operative computed tomography (CT) scans, taking into consideration the tumour sub-region that offers the greatest accuracy when grading. Furthermore, radiomics and machine learning were compared with biopsy-determined grading in a sub-group with resection histopathology as a reference standard. Background: Renal cancers are among the top ten causes of cancer-specific mortality, of which the ccRCC subtype is responsible for most cases. The grading of ccRCC is important in determining tumour aggressiveness and clinical management. Objectives: The objectives of this research were to predict the WHO/ISUP grade of ccRCC pre-operatively and characterise the heterogeneity of tumour sub-regions using radiomics and ML models, including comparison with pre-operative biopsy-determined grading in a sub-group. Methods: Data were obtained from multiple institutions across two countries, including 391 patients with pathologically proven ccRCC. For analysis, the data were separated into four cohorts. Cohorts 1 and 2 included data from the respective institutions from the two countries, cohort 3 was the combined data from both cohort 1 and 2, and cohort 4 was a subset of cohort 1, for which both the biopsy and subsequent histology from resection (partial or total nephrectomy) were available. 3D image segmentation was carried out to derive a voxel of interest (VOI) mask. Radiomics features were then extracted from the contrast-enhanced images, and the data were normalised. The Pearson correlation coefficient and the XGBoost model were used to reduce the dimensionality of the features. Thereafter, 11 ML algorithms were implemented for the purpose of predicting the ccRCC grade and characterising the heterogeneity of sub-regions in the tumours. Results: For cohort 1, the 50% tumour core and 25% tumour periphery exhibited the best performance, with an average AUC of 77.9% and 78.6%, respectively. The 50% tumour core presented the highest performance in cohorts 2 and 3, with average AUC values of 87.6% and 76.9%, respectively. With the 25% periphery, cohort 4 showed AUC values of 95.0% and 80.0% for grade prediction when using internal and external validation, respectively, while biopsy histology had an AUC of 31.0% for the classification with the final grade of resection histology as a reference standard. The CatBoost classifier was the best for each of the four cohorts with an average AUC of 80.0%, 86.5%, 77.0% and 90.3% for cohorts 1, 2, 3 and 4 respectively. Conclusions: Radiomics signatures combined with ML have the potential to predict the WHO/ISUP grade of ccRCC with superior performance, when compared to pre-operative biopsy. Moreover, tumour sub-regions contain useful information that should be analysed independently when determining the tumour grade. Therefore, it is possible to distinguish the grade of ccRCC pre-operatively to improve patient care and management. [ABSTRACT FROM AUTHOR]

Simple Summary: Surgery to prevent breast and ovarian cancer is very effective, but there is limited research into the quality of life afterwards, particularly with reference to quality-of-life scores known as utility scores. These are important for patients at increased cancer risk who are considering surgery and are essential for health-economic evaluations. This article used data from a systematic review of different quality-of-life questionnaires given to patients following these surgeries. We converted these data into utility scores, as recommended by the national guidelines. This shows that surgery to prevent breast cancer is associated with a long-term utility score of 0.92 and surgery to prevent ovarian cancer has a score of 0.97, indicating a mild–moderate impact. These are the first utility scores sourced from patients who have undergone these procedures, and they are important to doctors counselling patients about cancer prevention options and to researchers. Higher-quality studies are still needed, using the recommended quality-of-life questionnaire (EQ-5D). Background: Risk-reducing mastectomy (RRM) and risk-reducing salpingo-oophorectomy (RRSO) are the most effective breast and ovarian cancer preventive interventions. EQ-5D is the recommended tool to assess the quality of life and determine health-related utility scores (HRUSs), yet there are no published EQ-5D HRUSs after these procedures. These are essential for clinicians counselling patients and for health-economic evaluations. Methods: We used aggregate data from our published systematic review and converted SF-36/SF-12 summary scores to EQ-5D HRUSs using a published mapping algorithm. Study control arm or age-matched country-specific reference values provided comparison. Random-effects meta-analysis provided adjusted disutilities and utility scores. Subgroup analyses included long-term vs. short-term follow-up. Results: Four studies (209 patients) reported RRM outcomes using SF-36, and five studies (742 patients) reported RRSO outcomes using SF-12/SF-36. RRM is associated with a long-term (>2 years) disutility of −0.08 (95% CI −0.11, −0.04) (I2 31.4%) and a utility of 0.92 (95% CI 0.88, 0.95) (I2 31.4%). RRSO is associated with a long-term (>1 year) disutility of −0.03 (95% CI −0.05, 0.00) (I2 17.2%) and a utility of 0.97 (95% CI 0.94, 0.99) (I2 34.0%). Conclusions: We present the first HRUSs sourced from patients following RRM and RRSO. There is a need for high-quality prospective studies to characterise quality of life at different timepoints. [ABSTRACT FROM AUTHOR]

Simple Summary: Topic modeling, widely used in natural language processing, categorizes text documents into themes based on word frequency analysis. It has found success in various biological data analyses, including the accurate prediction of cancer subtypes and the simultaneous identification of genes, enhancers, and cell types from sparse single-cell data. Our study introduces a novel topic modeling approach for clustering single cells and detecting gene signatures in multi-omics single-cell datasets. Applied to study transcriptional heterogeneity in breast cancer cells resistant to chemotherapy and targeted therapy, it identifies protein-coding genes and long non-coding RNAs grouping cells into biologically similar clusters, effectively distinguishing between drug-sensitive and -resistant cancer types. Previous studies have interrogated long non-coding RNA (lncRNA) expression in single-cell data within breast cancer subtypes. Yet, the combined analysis of both lncRNA and mRNA expression in a cell type-specific manner remains to be explored. Compared to standard clustering methods, our approach offers a simultaneous optimal partitioning of genes and cells into topics and clusters, yielding easily interpretable results. Integrating mRNA and lncRNA data enhances cell classification accuracy. Topic modeling is a popular technique in machine learning and natural language processing, where a corpus of text documents is classified into themes or topics using word frequency analysis. This approach has proven successful in various biological data analysis applications, such as predicting cancer subtypes with high accuracy and identifying genes, enhancers, and stable cell types simultaneously from sparse single-cell epigenomics data. The advantage of using a topic model is that it not only serves as a clustering algorithm, but it can also explain clustering results by providing word probability distributions over topics. Our study proposes a novel topic modeling approach for clustering single cells and detecting topics (gene signatures) in single-cell datasets that measure multiple omics simultaneously. We applied this approach to examine the transcriptional heterogeneity of luminal and triple-negative breast cancer cells using patient-derived xenograft models with acquired resistance to chemotherapy and targeted therapy. Through this approach, we identified protein-coding genes and long non-coding RNAs (lncRNAs) that group thousands of cells into biologically similar clusters, accurately distinguishing drug-sensitive and -resistant breast cancer types. In comparison to standard state-of-the-art clustering analyses, our approach offers an optimal partitioning of genes into topics and cells into clusters simultaneously, producing easily interpretable clustering outcomes. Additionally, we demonstrate that an integrative clustering approach, which combines the information from mRNAs and lncRNAs treated as disjoint omics layers, enhances the accuracy of cell classification. [ABSTRACT FROM AUTHOR]

Simple Summary: Pancreatic neuroendocrine tumors are a rare and heterogenous group of neoplasms. Surgical resection is the only curative option. However, there has been an increase in palliative medical options. The aim of this retrospective study was to investigate responses for the most commonly used medical treatments in 192 patients. The current results support the effectiveness of somatostatin analogues in low-grade tumors and showed that it might also be used in patients with Ki-67 ≥ 10%. Treatment with streptozocin and 5-floururacil as first-line treatment showed good efficacy for G2 disease. Due to good efficacy and generally good tolerability PRRT might be considered as first-line treatment for NET G2. The results confirmed poor prognosis in high-grade tumors treated with carboplatin/etoposide or temozolomide. The current results provide valuable knowledge as current treatment algorithms and sequencing are primarily guided by expert opinions with limited evidence. Background: Given the rarity and heterogeneity of pancreatic neuroendocrine neoplasms (pNEN), treatment algorithms and sequencing are primarily guided by expert opinions with limited evidence. Aim: To investigate overall survival (OS), median progression-free survival (mPFS), and prognostic factors associated with the most common medical treatments for pNEN. Methods: Retrospective single-center study encompassing patients diagnosed and monitored between 2000 and 2020 (n = 192). Results: Median OS was 36 (95% CI: 26–46) months (99 months for grade (G) 1, 62 for G2, 14 for G3, and 10 for neuroendocrine carcinomas). Patients treated with somatostatin analogues (SSA) (n = 59, median Ki-67 9%) had an mPFS of 28 months. Treatment line (HR (first line as reference) 4.1, 95% CI: 1.9–9.1, p ≤ 0.001) emerged as an independent risk factor for time to progression. Patients with a Ki-67 index ≥10% (n = 28) had an mPFS of 27 months. Patients treated with streptozocin/5-fluorouracil (STZ/5FU) (n = 70, first-line treatment n = 68, median Ki-67 10%) had an mPFS of 20 months, with WHO grade serving as an independent risk factor (HR (G1 (n = 8) vs. G2 (n = 57)) 2.8, 95% CI: 1.1–7.2, p-value = 0.031). Median PFS was 21 months for peptide receptor radionuclide therapy (PRRT) (n = 41, first line n = 2, second line n = 29, median Ki-67 8%), 5 months for carboplatin and etoposide (n = 66, first-line treatment n = 60, median Ki-67 80%), and 3 months for temozolomide-based therapy (n = 56, first-line treatment n = 17, median Ki-67 30%). Conclusion: (1) Overall survival was, as expected, highly dependent on grade; (2) median PFS for SSA was around 2.5 years without difference between tumors with Ki-67 above or below 10%; (3) STZ/5FU as first-line treatment exhibited a superior mPFS of 20 months compared to what has historically been reported for targeted treatments; (4) PRRT in G2 pNEN achieved an mPFS similar to first-line chemotherapy; and (5) limited treatment efficacy was observed in high-grade tumors when treated with carboplatin and etoposide or temozolomide. [ABSTRACT FROM AUTHOR]

Simple Summary: Bronchopleural fistula after lobectomy for lung cancer is a rare and life-threatening complication. Recognizing and managing this complication in time is often very complex for the clinicians and surgeons. Several surgical and endoscopic therapeutic approaches were described for the management of broncho-pleural fistula during the time period under study; however, the various instituted treatments depend much more on the experiences of individual centers than on a codified algorithm. Indeed, at present, there are no codified guidelines in the literature for its treatment. On the basis of our 25 years of experience, we tried to outline a diagnostic and therapeutic algorithm to help clinicians in the management of this feared complication Backgrounds: Our goal is to evaluate the correct management of broncho-pleural fistula (BPF) after lobectomy for lung cancer. Methods: We retrospectively reviewed our 25-years' experience and reported our strategies and our diagnostic algorithm for the management of post-lobectomy broncho-pleural fistula. Results: Five thousand one hundred and fifty (5150) patients underwent lobectomy for lung cancer in the period between 1998 and 2023. A total of 44 (0.85%) out of 5150 developed post-operative BPF. In 11 cases, BPF was solved by non-invasive treatment. In nine cases, direct surgical repair of the bronchial stump allowed BPF resolution. In 14 cases, a completion intervention was performed. In six cases, we performed open window thoracostomy (OWT) after lobectomy; in two cases, the BPF was closed by percutaneous injection of an n-butyl cyanoacrylate glue mixture. In two cases, no surgical procedure was performed because of the clinical status of the patient at the time of fistula developing. Thirty-day and ninety-day mortality from fistula onset was, respectively, 18.2% (eight patients) and 22.7% (ten patients). Thirty-day and ninety-day mortality after completion pneumonectomy (12 patients) was, respectively, 8.3% (one patient) and 16.6% (two patients). Conclusions: The correct management of BPF depends on various factors: timing of onset, size of the fistula, anatomic localization, and the general condition of the patient. In the case of failure of various initial therapeutic approaches, completion intervention or OWT could be considered. [ABSTRACT FROM AUTHOR]

Simple Summary: Early characterization of the isocitrate dehydrogenase (IDHIDH) gene mutation status of glioma is crucial for personalized decision making and prognosis in clinical neurooncological treatment. Based on the known differences in energy metabolism between IDHIDH-mutated and IDHIDH-wildtype gliomas, we assessed physio-metabolic magnetic resonance imaging-based measures along with machine learning for potential reliable presurgical characterization of IDHIDH gene status. Traditional machine learning algorithms and simple deep learning models trained in analyzing physio-metabolic parameters demonstrated the best performance in classifying the IDHIDH gene status of gliomas in independent internal testing. In contrast, external testing revealed that traditional machine learning models trained on clinical MRI data had higher accuracy compared to physio-metabolic algorithms, reflecting differences in data acquisition methodology between the two sites. Our results outline the necessity of independent internal and external testing, thus calling for standardized and robust protocols for clinical MRI-driven AI applications. The mutational status of the isocitrate dehydrogenase (IDH) gene plays a key role in the treatment of glioma patients because it is known to affect energy metabolism pathways relevant to glioma. Physio-metabolic magnetic resonance imaging (MRI) enables the non-invasive analysis of oxygen metabolism and tissue hypoxia as well as associated neovascularization and microvascular architecture. However, evaluating such complex neuroimaging data requires computational support. Traditional machine learning algorithms and simple deep learning models were trained with radiomic features from clinical MRI (cMRI) or physio-metabolic MRI data. A total of 215 patients (first center: 166 participants + 16 participants for independent internal testing of the algorithms versus second site: 33 participants for independent external testing) were enrolled using two different physio-metabolic MRI protocols. The algorithms trained with physio-metabolic data demonstrated the best classification performance in independent internal testing: precision, 91.7%; accuracy, 87.5%; area under the receiver operating curve (AUROC), 0.979. In external testing, traditional machine learning models trained with cMRI data exhibited the best IDH classification results: precision, 84.9%; accuracy, 81.8%; and AUROC, 0.879. The poor performance for the physio-metabolic MRI approach appears to be explainable by site-dependent differences in data acquisition methodologies. The physio-metabolic MRI approach potentially supports reliable classification of IDH gene status in the presurgical stage of glioma patients. However, non-standardized protocols limit the level of evidence and underlie the need for a reproducible framework of data acquisition techniques. [ABSTRACT FROM AUTHOR]

Simple Summary: New technologies, particularly artificial intelligence (AI) and machine learning, enable the utilization of extensive data for personalized medicine. Future challenge involve acquiring limited biological data and extracting information for predictive models. This perspective focuses on triple-negative breast cancer (TNBC), a disease lacking specific treatment, where ongoing investigations explore therapies like immunotherapy. Drawing from the successful use of immunotherapy in melanoma, this perspective explores the identified strengths and weaknesses to inform more successful strategies. In this context, AI and predictive tools are invaluable. Advancements in melanoma immunotherapy serve as a foundation for developing effective immunotherapies for TNBC. Common principles, including immune system activation, checkpoint inhibitors, and personalized treatment, offer prospects for improving the outcomes in aggressive breast cancer cases, presenting new hope for challenging-to-treat patients, avoiding overtreatment, and reducing costs. Thanks to new technologies using artificial intelligence (AI) and machine learning, it is possible to use large amounts of data to try to extract information that can be used for personalized medicine. The great challenge of the future is, on the one hand, to acquire masses of biological data that nowadays are still limited and, on the other hand, to develop innovative strategies to extract information that can then be used for the development of predictive models. From this perspective, we discuss these aspects in the context of triple-negative breast cancer, a tumor where a specific treatment is still lacking and new therapies, such as immunotherapy, are under investigation. Since immunotherapy is already in use for other tumors such as melanoma, we discuss the strengths and weaknesses identified in the use of immunotherapy with melanoma to try to find more successful strategies. It is precisely in this context that AI and predictive tools can be extremely valuable. Therefore, the discoveries and advancements in immunotherapy for melanoma provide a foundation for developing effective immunotherapies for triple-negative breast cancer. Shared principles, such as immune system activation, checkpoint inhibitors, and personalized treatment, can be applied to TNBC to improve patient outcomes and offer new hope for those with aggressive, hard-to-treat breast cancer. [ABSTRACT FROM AUTHOR]

Simple Summary: Small-target classification in an image is still challenging in spite of emerging deep learning-based techniques. This study focused on the development of deep learning models for small-lesion classification. The proposed Rense block and edge conservative module enables deep learning models to extract better features of small lesions in images. Our RenseNet was validated with quantitative classification accuracy and a qualitative explanation heatmap for kidney stone and lung tumor computed tomography (CT) image datasets. Deep learning has become an essential tool in medical image analysis owing to its remarkable performance. Target classification and model interpretability are key applications of deep learning in medical image analysis, and hence many deep learning-based algorithms have emerged. Many existing deep learning-based algorithms include pooling operations, which are a type of subsampling used to enlarge the receptive field. However, pooling operations degrade the image details in terms of signal processing theory, which is significantly sensitive to small objects in an image. Therefore, in this study, we designed a Rense block and edge conservative module to effectively manipulate previous feature information in the feed-forward learning process. Specifically, a Rense block, an optimal design that incorporates skip connections of residual and dense blocks, was demonstrated through mathematical analysis. Furthermore, we avoid blurring of the features in the pooling operation through a compensation path in the edge conservative module. Two independent CT datasets of kidney stones and lung tumors, in which small lesions are often included in the images, were used to verify the proposed RenseNet. The results of the classification and explanation heatmaps show that the proposed RenseNet provides the best inference and interpretation compared to current state-of-the-art methods. The proposed RenseNet can significantly contribute to efficient diagnosis and treatment because it is effective for small lesions that might be misclassified or misinterpreted. [ABSTRACT FROM AUTHOR]

Simple Summary: The use of AI on medical images (CT, MRI, PET) has become a primary clinical and research interest. The main issues of these applications are strictly related to the reconstruction of imaging, the segmentation of tissues acquired, the selection of features, and the proper data analyses. Different approaches of AI have been proposed as the machine and deep learning, which utilize artificial neural networks inspired by neuronal architectures. Further validation of AI models for diagnosis and monitoring responses will be necessary to assess as MRI and PET/CT might provide a personalized treatment-response prediction superior to current methods. The lack of early detection and a high rate of recurrence/progression after surgery are defined as the most common causes of a very poor prognosis of Gliomas. The developments of quantification systems with special regards to artificial intelligence (AI) on medical images (CT, MRI, PET) are under evaluation in the clinical and research context in view of several applications providing different information related to the reconstruction of imaging, the segmentation of tissues acquired, the selection of features, and the proper data analyses. Different approaches of AI have been proposed as the machine and deep learning, which utilize artificial neural networks inspired by neuronal architectures. In addition, new systems have been developed using AI techniques to offer suggestions or make decisions in medical diagnosis, emulating the judgment of radiologist experts. The potential clinical role of AI focuses on the prediction of disease progression in more aggressive forms in gliomas, differential diagnosis (pseudoprogression vs. proper progression), and the follow-up of aggressive gliomas. This narrative Review will focus on the available applications of AI in brain tumor diagnosis, mainly related to malignant gliomas, with particular attention to the postoperative application of MRI and PET imaging, considering the current state of technical approach and the evaluation after treatment (including surgery, radiotherapy/chemotherapy, and prognostic stratification). [ABSTRACT FROM AUTHOR]

Simple Summary: Pancreatic ductal adenocarcinoma (PDAC) is often diagnosed at advanced stages, resulting in limited treatment options and poor survival rates. To address this challenge, we conducted a comprehensive analysis of pyroptosis-related genes using advanced algorithms. Our study, involving 1273 PDAC cases, identified 357 pyroptosis-related genes. Notably, BHLHE40, IL18, BIRC3, and APOL1 were found to be related to unfavourable PDAC outcomes and were validated through experiments and multiple datasets. We developed a novel model and an accessible nomogram to predict PDAC prognosis. Our research enhances our understanding of PDAC and has significant implications for both research and clinical practice. Pancreatic ductal adenocarcinoma (PDAC) is often diagnosed at late stages, limiting treatment options and survival rates. Pyroptosis-related gene signatures hold promise as PDAC prognostic markers, but limited gene pools and small sample sizes hinder their utility. We aimed to enhance PDAC prognosis with a comprehensive multi-algorithm analysis. Using R, we employed natural language processing and latent Dirichlet allocation on PubMed publications to identify pyroptosis-related genes. We collected PDAC transcriptome data (n = 1273) from various databases, conducted a meta-analysis, and performed differential gene expression analysis on tumour and non-cancerous tissues. Cox and LASSO algorithms were used for survival modelling, resulting in a pyroptosis-related gene expression-based prognostic index. Laboratory and external validations were conducted. Bibliometric analysis revealed that pyroptosis publications focus on signalling pathways, disease correlation, and prognosis. We identified 357 pyroptosis-related genes, validating the significance of BHLHE40, IL18, BIRC3, and APOL1. Elevated expression of these genes strongly correlated with poor PDAC prognosis and guided treatment strategies. Our accessible nomogram model aids in PDAC prognosis and treatment decisions. We established an improved gene signature for pyroptosis-related genes, offering a novel model and nomogram for enhanced PDAC prognosis. [ABSTRACT FROM AUTHOR]

Simple Summary: Immune-related adverse events (irAEs) are becoming an increasingly prevalent and well-studied phenomenon. Little is known about how clinical practice has evolved to incorporate our growing knowledge on the subject. We identified temporal trends in irAE management as well as factors associated with improved outcomes among these patients such as timely specialty consultation. These findings can improve the quality of management algorithms for immune-related adverse events at our institution and may inform policies in other institutions. Understanding of immune-related adverse events (irAEs) has evolved rapidly, and management guidelines are continually updated. We explored temporal changes in checkpoint inhibitor-induced irAE management at a tertiary cancer care center to identify areas for improvement. We conducted a single-center retrospective study of patients who developed a gastrointestinal, pulmonary, renal, or cardiac irAE between July and 1 October in 2019 or 2021. We collected patient demographic and clinical information up to 1 year after toxicity. Endoscopic evaluation and specialty follow-up after discharge for patients with gastrointestinal irAEs declined between the 2019 and 2021 periods. Symptom duration and steroid taper attempts also declined. For pulmonary irAEs, rates of specialty consultation, hospital admission and readmission, and mortality improved in 2021 compared with 2019. Follow-up rates after hospital discharge were consistently low (<50%) in both periods. For cardiac irAEs, consultation with a cardiologist was frequent and prompt in both periods. Outpatient treatment and earlier specialty consultation improved outcomes with gastrointestinal irAEs. Our study exploring irAE practice changes over time identified areas to improve management; specifically, timely specialty consultation was associated with better outcomes for gastrointestinal irAEs. These findings can help improve the quality of management algorithms at our institution and may inform policies in other institutions. [ABSTRACT FROM AUTHOR]

Simple Summary: This study proposes an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images, which is crucial for fully automated immunohistochemistry quantification. The proposed method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional staining slides. The model demonstrated an impressive intersection over union score on the test set, and a fully automated Ki-67 scoring system based on the model's predictions exhibited high consistency with the scores given by experienced pathologists. The proposed method brings the breast cancer fully immunohistochemistry quantitative scoring system one step closer to clinical application. Aims: The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions. Methods and Results: In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists. Conclusions: Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system. [ABSTRACT FROM AUTHOR]

Simple Summary: Tumor cells shed pieces of DNA that circulate in biofluids like blood and urine. The amount of circulating tumor DNA is related to the amount of cancer in the body. Knowing the amount of cancer helps doctors predict outcomes and decide treatments for patients. Tumor DNA has methylation patterns that makes it distinct from non-cancer DNA. In this study, we describe a new method called tumor methylated fraction (TMeF) that quantifies the cancer-indicative methylation patterns within circulating tumor DNA from blood samples. This method is non-invasive because it does not need tumor tissue to estimate the amount of cancer in the body. In the future, doctors could use TMeF to supplement current cancer screening methods. Estimating the abundance of cell-free DNA (cfDNA) fragments shed from a tumor (i.e., circulating tumor DNA (ctDNA)) can approximate tumor burden, which has numerous clinical applications. We derived a novel, broadly applicable statistical method to quantify cancer-indicative methylation patterns within cfDNA to estimate ctDNA abundance, even at low levels. Our algorithm identified differentially methylated regions (DMRs) between a reference database of cancer tissue biopsy samples and cfDNA from individuals without cancer. Then, without utilizing matched tissue biopsy, counts of fragments matching the cancer-indicative hyper/hypo-methylated patterns within DMRs were used to determine a tumor methylated fraction (TMeF; a methylation-based quantification of the circulating tumor allele fraction and estimate of ctDNA abundance) for plasma samples. TMeF and small variant allele fraction (SVAF) estimates of the same cancer plasma samples were correlated (Spearman's correlation coefficient: 0.73), and synthetic dilutions to expected TMeF of 10−3 and 10−4 had estimated TMeF within two-fold for 95% and 77% of samples, respectively. TMeF increased with cancer stage and tumor size and inversely correlated with survival probability. Therefore, tumor-derived fragments in the cfDNA of patients with cancer can be leveraged to estimate ctDNA abundance without the need for a tumor biopsy, which may provide non-invasive clinical approximations of tumor burden. [ABSTRACT FROM AUTHOR]