Search

The assessment of vascular complexity in the lower limbs provides relevant information about peripheral artery occlusive diseases (PAOD), thus fostering improvements both in therapeutic decisions and prognostic estimation. The current clinical practice consists of visually inspecting and evaluating cine-angiograms of the interested region, which is largely operator-dependent. We present here an automatic method for segmenting the vessel tree and compute a quantitative measure, in terms of fractal dimension (FD), of the vascular complexity. The proposed workflow consists of three main steps: (i) conversion of the cine-angiographies to single static images with a broader field of view, (ii) automatic segmentation of the vascular trees, and (iii) calculation and assessment of FD as complexity index. In particular, this work defines (1) a method to reduce the inter-observer variability in judging vascular complexity in cine-angiography images from patients affected by peripheral artery occlusive disease (PAOD), and (2) the use of Fractal Dimension as a metric of shape complexity of vascular tree. The inter-class correlation coefficient (ICC) is computed as inter-observer agreement metric and to account for possible systematic error, that depends on the experience of the raters. The automatic segmentation of vascular tree achieved an Area Under the Curve mean value of $$0.77~\pm ~0.07$$ 0.77 ± 0.07 , with a min-max range of $$0.57-0.87$$ 0.57 - 0.87 . Absolute operator agreement was higher over the segmented image ($$ICC=0.96$$ I C C = 0.96 ) compared to the video ($$ICC=0.76$$ I C C = 0.76 ) and the a broader field of view image ($$ICC=0.92$$ I C C = 0.92 ). Fractal Dimension computed on both manual segmented images (ground truths) and automatically showed a good correlation with the clinical score (0.85 and 0.75, respectively). Experimental analyses suggest that extracting the vascular tree from cine-angiography can substantially improve the reliability of visual assessment of vascular complexity in PAOD. Results also reveal the effectiveness of FD in evaluating complex vascular tree structures.

Recently, deep learning (DL)-based methods for the generation of synthetic computed tomography (sCT) have received significant research attention as an alternative to classical ones. We present here a systematic review of these methods by grouping them into three categories, according to their clinical applications: I) To replace CT in magnetic resonance (MR)-based treatment planning. II) Facilitate cone-beam computed tomography (CBCT)-based image-guided adaptive radiotherapy. III) Derive attenuation maps for the correction of positron emission tomography (PET). Appropriate database searching was performed on journal articles published between January 2014 and December 2020. The DL methods' key characteristics were extracted from each eligible study, and a comprehensive comparison among network architectures and metrics was reported. A detailed review of each category was given, highlighting essential contributions, identifying specific challenges, and summarising the achievements. Lastly, the statistics of all the cited works from various aspects were analysed, revealing the popularity and future trends, and the potential of DL-based sCT generation. The current status of DL-based sCT generation was evaluated, assessing the clinical readiness of the presented methods., 59 pages, 4 figures, 5 tables; two first equally contributing authors; Medical Physics. 2021

Cone-beam computed tomography (CBCT)- and magnetic resonance (MR)-images allow a daily observation of patient anatomy but are not directly suited for accurate proton dose calculations. This can be overcome by creating synthetic CTs (sCT) using deep convolutional neural networks. In this study, we compared sCTs based on CBCTs and MRs for head and neck (H&N) cancer patients in terms of image quality and proton dose calculation accuracy. A dataset of 27 H&N-patients, treated with proton therapy (PT), containing planning CTs (pCTs), repeat CTs, CBCTs and MRs were used to train two neural networks to convert either CBCTs or MRs into sCTs. Image quality was quantified by calculating mean absolute error (MAE), mean error (ME) and Dice similarity coefficient (DSC) for bones. The dose evaluation consisted of a systematic non-clinical analysis and a clinical recalculation of actually used proton treatment plans. Gamma analysis was performed for non-clinical and clinical treatment plans. For clinical treatment plans also dose to targets and organs at risk (OARs) and normal tissue complication probabilities (NTCP) were compared. CBCT-based sCTs resulted in higher image quality with an average MAE of 40 ± 4 HU and a DSC of 0.95, while for MR-based sCTs a MAE of 65 ± 4 HU and a DSC of 0.89 was observed. Also in clinical proton dose calculations, sCTCBCT achieved higher average gamma pass ratios (2%/2 mm criteria) than sCTMR (96.1% vs. 93.3%). Dose-volume histograms for selected OARs and NTCP-values showed a very small difference between sCTCBCT and sCTMR and a high agreement with the reference pCT. CBCT- and MR-based sCTs have the potential to enable accurate proton dose calculations valuable for daily adaptive PT. Significant image quality differences were observed but did not affect proton dose calculation accuracy in a similar manner. Especially the recalculation of clinical treatment plans showed high agreement with the pCT for both sCTCBCT and sCTMR.

Coronary Angiography (CA) is the standard of reference to diagnose coronary artery disease. Yet, only a portion of the information it conveys is usually used. Quantitative Coronary Angiography (QCA) reliably contributes to improving the measurable assessment of CA. In this work, we developed a new software, CoroFinder, able to automatically identify epicardial coronary arteries and to dynamically track the vessel profile in dye-free frames. The coronary tree is automatically segmented by Frangi’s filter in the angiogram’s frames where vessels are contrasted (“template frames”). Afterward, the image similarity among each template frame and the dye-free images is scored by cross-correlation. Finally, each dye-free image is associated with the most similar template frame, resulting in an estimation of vessel contour. CoroFinder allows locating the position of coronary arteries in absence of contrast dye. The developed algorithm is robust to diverse vessel curvatures, variation of vessel widths, and the presence of stenoses. This article describes the newly developed CoroFinder algorithm and the associated software and provides an overview of its potential application in research and for translation to the clinic.

With the advent of Minimally Invasive Surgery (MIS), intra-operative imaging has become crucial for surgery and therapy guidance, allowing to partially compensate for the lack of information typical of MIS. This paper reviews the advancements in both classical (i.e. ultrasounds, X-ray, optical coherence tomography and magnetic resonance imaging) and more recent (i.e. multispectral, photoacoustic and Raman imaging) intra-operative imaging modalities. Each imaging modality was analyzed, focusing on benefits and disadvantages in terms of compatibility with the operating room, costs, acquisition time and image characteristics. Tables are included to summarize this information. New generation of hybrid surgical room and algorithms for real time/in room image processing were also investigated. Each imaging modality has its own (site- and procedure-specific) peculiarities in terms of spatial and temporal resolution, field of view and contrasted tissues. Besides the benefits that each technique offers for guidance, considerations about operators and patient risk, costs, and extra time required for surgical procedures have to be considered. The current trend is to equip surgical rooms with multimodal imaging systems, so as to integrate multiple information for real-time data extraction and computer-assisted processing. The future of surgery is to enhance surgeons eye to minimize intra- and after-surgery adverse events and provide surgeons with all possible support to objectify and optimize the care-delivery process.

Background Time-resolved 4D cone beam-computed tomography (4D-CBCT) allows a daily assessment of patient anatomy and respiratory motion. However, 4D-CBCTs suffer from imaging artifacts that affect the CT number accuracy and prevent accurate proton dose calculations. Deep learning can be used to correct CT numbers and generate synthetic CTs (sCTs) that can enable CBCT-based proton dose calculations. Purpose In this work, sparse view 4D-CBCTs were converted into 4D-sCT utilizing a deep convolutional neural network (DCNN). 4D-sCTs were evaluated in terms of image quality and dosimetric accuracy to determine if accurate proton dose calculations for adaptive proton therapy workflows of lung cancer patients are feasible. Methods A dataset of 45 thoracic cancer patients was utilized to train and evaluate a DCNN to generate 4D-sCTs, based on sparse view 4D-CBCTs reconstructed from projections acquired with a 3D acquisition protocol. Mean absolute error (MAE) and mean error were used as metrics to evaluate the image quality of single phases and average 4D-sCTs against 4D-CTs acquired on the same day. The dosimetric accuracy was checked globally (gamma analysis) and locally for target volumes and organs-at-risk (OARs) (lung, heart, and esophagus). Furthermore, 4D-sCTs were also compared to 3D-sCTs. To evaluate CT number accuracy, proton radiography simulations in 4D-sCT and 4D-CTs were compared in terms of range errors. The clinical suitability of 4D-sCTs was demonstrated by performing a 4D dose reconstruction using patient specific treatment delivery log files and breathing signals. Results 4D-sCTs resulted in average MAEs of 48.1 +/- 6.5 HU (single phase) and 37.7 +/- 6.2 HU (average). The global dosimetric evaluation showed gamma pass ratios of 92.3% +/- 3.2% (single phase) and 94.4% +/- 2.1% (average). The clinical target volume showed high agreement in D-98 between 4D-CT and 4D-sCT, with differences below 2.4% for all patients. Larger dose differences were observed in mean doses of OARs (up to 8.4%). The comparison with 3D-sCTs showed no substantial image quality and dosimetric differences for the 4D-sCT average. Individual 4D-sCT phases showed slightly lower dosimetric accuracy. The range error evaluation revealed that lung tissues cause range errors about three times higher than the other tissues. Conclusion In this study, we have investigated the accuracy of deep learning-based 4D-sCTs for daily dose calculations in adaptive proton therapy. Despite image quality differences between 4D-sCTs and 3D-sCTs, comparable dosimetric accuracy was observed globally and locally. Further improvement of 3D and 4D lung sCTs could be achieved by increasing CT number accuracy in lung tissues.

Microarray is a powerful technology that enables the monitoring of expression levels for thousands of genes simultaneously, providing scientists with a full overview about DNA and RNA investigation. The process is made of three main phases: interaction with biological samples, data extraction, and data analysis. In particular, the data extraction phase strongly relies on image processing algorithms, since the expression levels are revealed by the interaction of light with fluorescent markers. More in detail, in order to extract quantitative information from probes image, three steps are required: (1) gridding, (2) segmentation, and (3) intensity quantification. Errors in one of these steps can deeply affect the process outcome. In this chapter each of the above mentioned steps will be analyzed and discussed. Software platforms dedicated to this purpose will be reported as well.

PURPOSE: Adaptive proton therapy (APT) of lung cancer patients requires frequent volumetric imaging of diagnostic quality. Cone-beam CT (CBCT) can provide these daily images, but x-ray scattering limits CBCT-image quality and hampers dose calculation accuracy. The purpose of this study was to generate CBCT-based synthetic CTs using a deep convolutional neural network (DCNN) and investigate image quality and clinical suitability for proton dose calculations in lung cancer patients.METHODS: A dataset of 33 thoracic cancer patients, containing CBCTs, same-day repeat CTs (rCT), planning-CTs (pCTs) and clinical proton treatment plans, was used to train and evaluate a DCNN with and without a pCT-based correction method. Mean absolute error (MAE), mean error (ME), peak signal-to-noise ratio and structural similarity were used to quantify image quality. The evaluation of clinical suitability was based on recalculation of clinical proton treatment plans. Gamma pass ratios, mean dose to target volumes and organs at risk, and normal tissue complication probabilities (NTCP) were calculated. Furthermore, proton radiography simulations were performed to assess the HU-accuracy of sCTs in terms of range errors.RESULTS: On average, sCTs without correction resulted in a MAE of 34±6 HU and ME of 4±8 HU. The correction reduced the MAE to 31±4HU (ME to 2±4HU). Average 3%/3mm gamma pass ratios increased from 93.7% to 96.8%, when the correction was applied. The patient specific correction reduced mean proton range errors from 1.5 to 1.1 mm. Relative mean target dose differences between sCTs and rCT were below ±0.5% for all patients and both synthetic CTs (with/without correction). NTCP values showed high agreement between sCTs and rCT (CONCLUSION: CBCT-based sCTs can enable accurate proton dose calculations for APT of lung cancer patients. The patient specific correction method increased the image quality and dosimetric accuracy but had only a limited influence on clinically relevant parameters. This article is protected by copyright. All rights reserved.

The coronavirus disease 19 (COVID-19) pandemic is having a dramatic impact on society and healthcare systems. In this complex scenario, lung computerized tomography (CT) may play an important prognostic role. However, datasets released so far present limitations that hamper the development of tools for quantitative analysis. In this paper, we present an open-source lung CT dataset comprising information on 50 COVID-19-positive patients. The CT volumes are provided along with (i) an automatic threshold-based annotation obtained with a Gaussian mixture model (GMM) and (ii) a scoring provided by an expert radiologist. This score was found to significantly correlate with the presence of ground glass opacities and the consolidation found with GMM. The dataset is freely available in an ITK-based file format under the CC BY-NC 4.0 license. The code for GMM fitting is publicly available, as well. We believe that our dataset will provide a unique opportunity for researchers working in the field of medical image analysis, and hope that its release will lay the foundations for the successfully implementation of algorithms to support clinicians in facing the COVID-19 pandemic.

Background: This study aims to examine the underlying associations between eating, affective and metacognitive symptoms in patients with binge eating disorder (BED) through network analysis (NA), in order to identify key variables that may be considered the target for psychotherapeutic interventions. Methods: One hundred and fifty-five patients with BED completed measures of eating psychopathology, affective symptoms, emotion regulation and metacognition. A cross-sectional network was inferred by means of Gaussian Markov random field estimation using graphical LASSO and extended Bayesian information criterion (EBIC-LASSO), and central symptoms of BED were identified by means of the strength centrality index. Results: Impaired self-monitoring metacognition and difficulties on impulse control emerged as the symptoms with the highest centrality. Conversely, eating and affective features were less central. The centrality stability coefficient of strength was above the recommended cut-off, thus indicating the stability of the network. Conclusions: According to present NA findings, impaired self-monitoring metacognition and difficulties on impulse control are the central nodes in the psychopathological network of BED while eating symptoms appear marginal. If further studies with larger samples replicate these results, metacognition and impulse control could represent new targets of psychotherapeutic interventions in the treatment of BED. In light of this, Metacognitive Interpersonal Therapy (MIT) could be a promising aid in clinical practice to develop an effective treatment for BED.

In-room imaging is a prerequisite for adaptive proton therapy. The use of onboard cone-beam computed tomography (CBCT) imaging, which is routinely acquired for patient position verification, can enable daily dose reconstructions and plan adaptation decisions. Image quality deficiencies though, hamper dose calculation accuracy and make corrections of CBCTs a necessity. This study compared three methods to correct CBCTs and create synthetic CTs that are suitable for proton dose calculations. CBCTs, planning CTs and repeated CTs (rCT) from 33 H&N cancer patients were used to compare a deep convolutional neural network (DCNN), deformable image registration (DIR) and an analytical image-based correction method (AIC) for synthetic CT (sCT) generation. Image quality of sCTs was evaluated by comparison with a same-day rCT, using mean absolute error (MAE), mean error (ME), Dice similarity coefficient (DSC), structural non-uniformity (SNU) and signal/contrast-to-noise ratios (SNR/CNR) as metrics. Dosimetric accuracy was investigated in an intracranial setting by performing gamma analysis and calculating range shifts. Neural network-based sCTs resulted in the lowest MAE and ME (37/2 HU) and the highest DSC (0.96). While DIR and AIC generated images with a MAE of 44/77 HU, a ME of −8/1 HU and a DSC of 0.94/0.90. Gamma and range shift analysis showed almost no dosimetric difference between DCNN and DIR based sCTs. The lower image quality of AIC based sCTs affected dosimetric accuracy and resulted in lower pass ratios and higher range shifts. Patient-specific differences highlighted the advantages and disadvantages of each method. For the set of patients, the DCNN created synthetic CTs with the highest image quality. Accurate proton dose calculations were achieved by both DCNN and DIR based sCTs. The AIC method resulted in lower image quality and dose calculation accuracy was reduced compared to the other methods.

Radiation therapy is one of the most important strategies for treating patients with tumor. The rationale is to deliver high radiation doses to the tumor in order to damage its DNA while sparing, at the same time, healthy tissues. In order to optimize such a process, biomedical images play a fundamental role; in particular, Magnetic Resonance (MR) produces well-contrasted images for precisely contouring tumors and organs at risk. However, due to the physical information stored in it, Computed Tomography (CT) is mandatory for accurate radiation dose calculation. To overcome this limitation, several algorithms, usually based on deep learning techniques, were proposed for converting MR–to–synthetic CT (sCT). In this paper, we report about the evaluation of the impact of three different train losses, commonly used for non–medical applications. Tests were ran on a cohort of 15 brain MR/CT image pairs. An algorithm for MR–to–sCT conversion, previously developed for MR-only radiotherapy, was used as benchmark platform. Predicted sCT images were compared on the basis of intensities and edges reconstruction. Results show that potential improvements can be achieved if non–medical image loss will be adapted to this application.

Purpose Automated delineation of structures and organs is a key step in medical imaging. However, due to the large number and diversity of structures and the large variety of segmentation algorithms, a consensus is lacking as to which automated segmentation method works best for certain applications. Segmentation challenges are a good approach for unbiased evaluation and comparison of segmentation algorithms. Methods In this work, we describe and present the results of the Head and Neck Auto-Segmentation Challenge 2015, a satellite event at the Medical Image Computing and Computer Assisted Interventions (MICCAI) 2015 conference. Six teams participated in a challenge to segment nine structures in the head and neck region of CT images: brainstem, mandible, chiasm, bilateral optic nerves, bilateral parotid glands, and bilateral submandibular glands. Results This paper presents the quantitative results of this challenge using multiple established error metrics and a well-defined ranking system. The strengths and weaknesses of the different auto-segmentation approaches are analyzed and discussed. Conclusions The Head and Neck Auto-Segmentation Challenge 2015 was a good opportunity to assess the current state-of-the-art in segmentation of organs at risk for radiotherapy treatment. Participating teams had the possibility to compare their approaches to other methods under unbiased and standardized circumstances. The results demonstrate a clear tendency toward more general purpose and fewer structure-specific segmentation algorithms.

Atlas-based automatic segmentation (ABAS) addresses the challenges of accuracy and reliability in manual segmentation. We aim to evaluate the contribution of specific-purpose in ABAS of breast cancer (BC) patients with respect to generic-purpose libraries.One generic-purpose and 9 specific-purpose libraries, stratified according to type of surgery and size of thorax circumference, were obtained from the computed tomography of 200 BC patients. Keywords about contralateral breast volume and presence of breast expander/prostheses were recorded. ABAS was validated on 47 independent patients, considering manual segmentation from scratch as reference. Five ABAS datasets were obtained, testing single-ABAS and multi-ABAS with simultaneous truth and performance level estimation (STAPLE). Center of mass distance (CMD), average Hausdorff distance (AHD) and Dice similarity coefficient (DSC) between corresponding ABAS and manual structures were evaluated and statistically significant differences between different surgeries, structures and ABAS strategies were investigated.Statistically significant differences between patients who underwent different surgery were found, with superior results for conservative-surgery group, and between different structures were observed: ABAS of heart, lungs, kidneys and liver was satisfactory (median values: CMD2 mm, DSC≥0.80, AHD1.5 mm), whereas chest wall, breast and spinal cord obtained moderate performance (median values: 2 mm ≤ CMD5 mm, 0.60 ≤ DSC0.80, 1.5 mm ≤ AHD4 mm) and esophagus, stomach, brachial plexus and supraclavicular nodes obtained poor performance (median CMD≥5 mm, DSC0.60, AHD≥4 mm). The application of STAPLE algorithm generally yields higher performance and the use of keywords improves results for breast ABAS.The homogeneity in the selection of atlases based on multiple anatomical and clinical features and the use of specific-purpose libraries can improve ABAS performance with respect to generic-purpose libraries.

The data of literature are discordant about the role of mast cells in different types of neoplasms. In this paper the authors propose the hypothesis that tumor-associated mast cells may switch to different polarization states, conditioning the immunogenic capacities of the different neoplasms. Anti-inflammatory polarized mast cells should express cytokines such as interleukin-10 (IL-10) and then mast cells number should be inversely related to the intensity of inflammatory infiltrate. On the contrary, when mast cells do not express anti-inflammatory cytokines their number should be directly related to the intensity of the inflammatory infiltrate. In this paper we briefly argue around feasible approaches, based on the retrospective studies of tumor tissue samples from neoplasms considered "immunologically hot" and neoplasms considered "immunologically cold", through immunohistochemistry and immunofluorescence techniques (confocal microscopy). The establishment of the actual existence of a polarization interchange of mast cells, could lead to a new vision in prognostic terms, useful to contrive new approaches in immunotherapy of tumors.

Purpose: Multiatlas based segmentation is largely used in many clinical and research applications. Due to its good performances, it has recently been included in some commercial platforms for radiotherapy planning and surgery guidance. Anyway, to date, a software with no restrictions about the anatomical district and image modality is still missing. In this paper we introduce plastimatch mabs, an open source software that can be used with any image modality for automatic segmentation. Methods: plastimatch mabs workflow consists of two main parts: (1) an offline phase, where optimal registration and voting parameters are tuned and (2) an online phase, where a new patient is labeled from scratch by using the same parameters as identified in the former phase. Several registration strategies, as well as different voting criteria can be selected. A flexible atlas selection scheme is also available. To prove the effectiveness of the proposed software across anatomical districts and image modalities, it was tested on two very different scenarios: head and neck (H&N) CT segmentation for radiotherapy application, and magnetic resonance image brain labeling for neuroscience investigation. Results: For the neurological study, minimum dice was equal to 0.76 (investigated structures: left and right caudate, putamen, thalamus, and hippocampus). For head and neck case, minimum dice was 0.42 for the most challenging structures (optic nerves and submandibular glands) and 0.62 for the other ones (mandible, brainstem, and parotid glands). Time required to obtain the labels was compatible with a real clinical workflow (35 and 120 min). Conclusions: The proposed software fills a gap in the multiatlas based segmentation field, since all currently available tools (both for commercial and for research purposes) are restricted to a well specified application. Furthermore, it can be adopted as a platform for exploring MABS parameters and as a reference implementation for comparing against other segmentation algorithms.

Interaction between medical image platform and external environment is a desirable feature in several clinical, research, and educational scenarios. In this work, the integration between 3D Slicer package and Arduino board is introduced, enabling a simple and useful communication between the two software/hardware platforms. The open source extension, programmed in Python language, manages the connection process and offers a communication layer accessible from any point of the medical image suite infrastructure. Deep integration with 3D Slicer code environment is provided and a basic input&ndash, output mechanism accessible via GUI is also made available. To test the proposed extension, two exemplary use cases were implemented: (1) INPUT data to 3D Slicer, to navigate on basis of data detected by a distance sensor connected to the board, and (2) OUTPUT data from 3D Slicer, to control a servomotor on the basis of data computed through image process procedures. Both goals were achieved and quasi-real-time control was obtained without any lag or freeze, thus boosting the integration between 3D Slicer and Arduino. This integration can be easily obtained through the execution of few lines of Python code. In conclusion, SlicerArduino proved to be suitable for fast prototyping, basic input&ndash, output interaction, and educational purposes. The extension is not intended for mission-critical clinical tasks.

The intricate relationships between innate immunity and brain diseases raise increased interest across the wide spectrum of neurodegenerative and neuropsychiatric disorders. Barriers, such as the blood−brain barrier, and innate immunity cells such as microglia, astrocytes, macrophages, and mast cells are involved in triggering disease events in these groups, through the action of many different cytokines. Chronic inflammation can lead to dysfunctions in large-scale brain networks. Neurodegenerative diseases, such as Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, amyotrophic lateral sclerosis, and frontotemporal dementia, are associated with a substrate of dysregulated immune responses that impair the central nervous system balance. Recent evidence suggests that similar phenomena are involved in psychiatric diseases, such as depression, schizophrenia, autism spectrum disorders, and post-traumatic stress disorder. The present review summarizes and discusses the main evidence linking the innate immunological response in neurodegenerative and psychiatric diseases, thus providing insights into how the responses of innate immunity represent a common denominator between diseases belonging to the neurological and psychiatric sphere. Improved knowledge of such immunological aspects could provide the framework for the future development of new diagnostic and therapeutic approaches.

The first aim of this work is to present a novel deep convolution neural network (DCNN) multiplane approach and compare it to single-plane prediction of synthetic computed tomography (sCT) by using the real computed tomography (CT) as ground truth. The second aim is to demonstrate the feasibility of magnetic resonance imaging (MRI)-based proton therapy planning for the brain by assessing the range shift error within the clinical acceptance threshold.The image database included 15 pairs of MRI/CT scans of the head. Three DCNNs were trained to estimate, for each voxel, the Hounsfield unit (HU) value from MRI intensities. Each DCNN gave an estimation in the axial, sagittal, and coronal plane, respectively. The median HU among the 3 values was selected to build the sCT. The sCT/CT agreement was evaluated by a mean absolute error (MAE) and mean error, computed within the head contour and on 6 different tissues. Dice similarity coefficients were calculated to assess the geometric overlap of bone and air cavities segmentations. A 3-beam proton therapy plan was simulated for each patient. Beam-by-beam range shift (RS) analysis was conducted to assess the proton-stopping power estimation. RS analysis was performed using clinically accepted thresholds of (1) 3.5% + 1 mm and (2) 2.5% + 1.5 mm of the total range.DCNN multiplane statistically outperformed single-plane prediction of sCT (P.025). MAE and mean error within the head were 54 ± 7 HU and -4 ± 17 HU (mean ± standard deviation), respectively. Soft tissues were very close to perfect agreement (11 ± 3 HU in terms of MAE). Segmentation of air and bone regions led to a Dice similarity coefficient of 0.92 ± 0.03 and 0.93 ± 0.02, respectively. Proton RS was always below clinical acceptance thresholds, with a relative RS error of 0.14% ± 1.11%.The multiplane DCNN approach significantly improved the sCT prediction compared with other DCNN methods presented in the literature. The method was demonstrated to be highly accurate for MRI-only proton planning purposes.

Recent studies have clarified many still unknown aspects related to innate immunity and the blood-brain barrier relationship. They have also confirmed the close links between effector immune system cells, such as granulocytes, macrophages, microglia, natural killer cells and mast cells, and barrier functionality. The latter, in turn, is able to influence not only the entry of the cells of the immune system into the nervous tissue, but also their own activation. Interestingly, these two components and their interactions play a role of great importance not only in infectious diseases, but in almost all the pathologies of the central nervous system. In this paper, we review the main aspects in the field of vascular diseases (cerebral ischemia), of primitive and secondary neoplasms of Central Nervous System CNS, of CNS infectious diseases, of most common neurodegenerative diseases, in epilepsy and in demyelinating diseases (multiple sclerosis). Neuroinflammation phenomena are constantly present in all diseases; in every different pathological state, a variety of innate immunity cells responds to specific stimuli, differentiating their action, which can influence the blood-brain barrier permeability. This, in turn, undergoes anatomical and functional modifications, allowing the stabilization or the progression of the pathological processes.

In radiation therapy, a key step for a successful cancer treatment is image-based treatment planning. One objective of the planning phase is the fast and accurate segmentation of organs at risk and target structures from medical images. However, manual delineation of organs, which is still the gold standard in many clinical environments, is time-consuming and prone to inter-observer variations. Consequently, many automated segmentation methods have been developed. In this work, we train two hierarchical 3D neural networks to segment multiple organs at risk in the head and neck area. First, we train a coarse network on size-reduced medical images to locate the organs of interest. Second, a subsequent fine network on full-resolution images is trained for a final accurate segmentation. The proposed method is purely deep learning based; accordingly, no pre-registration or post-processing is required. The approach has been applied on a publicly available computed tomography dataset, created for the MICCAI 2015 Auto-Segmentation challenge. In an extensive evaluation process, the best configurations for the trained networks have been determined. Compared to the existing methods, the presented approach shows state-of-the-art performance for the segmentation of seven different structures in the head and neck area. We conclude that 3D neural networks outperform the most existing model- and atlas-based methods for the segmentation of organs at risk in the head and neck area. The ease of use, high accuracy and the test time efficiency of the method make it promising for image-based treatment planning in clinical practice.

The assessment of vascular complexity in the lower limbs provides important information about peripheral artery diseases, with a relevant impact on both therapeutic decisions and on prognostic estimation. Currently, the evaluation is carried out by visual inspection of cine-angiograms, which is largely operator-dependent. An automatic image analysis could offer a fast and more reliable technique to support physicians with the clinical management of these patients. In this work, we introduce a new method to automatically segment the vascular tree from cine-angiography images, in order to improve the clinical interpretation of the complexity of vascular collaterals in Peripheral Arterial Occlusive Disease (PAOD) patients. The approach is based on: (1) a feature-detection method to convert the video into a static image with lager Field Of View (FOV) and (2) a custom Convolutional Neural Network (CNN) for the segmentation of vascular structure. Experimental evaluations over a set of clinical cases confirm the viability of the approach: accuracy is assessed in terms of area under the ROC curve, where an average value of \(0.988 \pm 0.006\) is measured.

Multi atlas based segmentation (MABS) uses a database of atlas images, and an atlas selection process is used to choose an atlas subset for registration and voting. In the current state of the art, atlases are chosen according to a similarity criterion between the target subject and each atlas in the database. In this paper, we propose a new concept for atlas selection that relies on selecting the best performing group of atlases rather than the group of highest scoring individual atlases. Experiments were performed using CT images of 50 patients, with contours of brainstem and parotid glands. The dataset was randomly split into two groups: 20 volumes were used as an atlas database and 30 served as target subjects for testing. Classic oracle selection, where atlases are chosen by the highest dice similarity coefficient (DSC) with the target, was performed. This was compared to oracle group selection, where all the combinations of atlas subgroups were considered and scored by computing DSC with the target subject. Subsequently, convolutional neural networks were designed to predict the best group of atlases. The results were also compared with the selection strategy based on normalized mutual information (NMI). Oracle group was proven to be significantly better than classic oracle selection (p

External-beam radiotherapy followed by high dose rate (HDR) brachytherapy is the standard-of-care for treating gynecologic cancers. The enhanced soft-tissue contrast provided by magnetic resonance imaging (MRI) makes it a valuable imaging modality for diagnosing and treating these cancers. However, in contrast to computed tomography (CT) imaging, the appearance of the brachytherapy catheters, through which radiation sources are inserted to reach the cancerous tissue later on, is often variable across images. This paper reports, for the first time, a new deep-learning-based method for fully automatic segmentation of multiple closely spaced brachytherapy catheters in intraoperative MRI. Represented in the data are 50 gynecologic cancer patients treated by MRI-guided HDR brachytherapy. For each patient, a single intraoperative MRI was used. 826 catheters in the images were manually segmented by an expert radiation physicist who is also a trained radiation oncologist. The number of catheters in a patient ranged between 10 and 35. A deep 3D convolutional neural network (CNN) model was developed and trained. In order to make the learning process more robust, the network was trained 5 times, each time using a different combination of shown patients. Finally, each test case was processed by the five networks and the final segmentation was generated by voting on the obtained five candidate segmentations. 4-fold validation was executed and all the patients were segmented. An average distance error of 2.0 +/- 3.4 mm was achieved. False positive and false negative catheters were 6.7% and 1.5% respectively. Average Dice score was equal to 0.60 +/- 0.17. The algorithm is available for use in the open source software platform 3D Slicer allowing for wide scale testing and research discussion. In conclusion, to the best of our knowledge, fully automatic segmentation of multiple closely spaced catheters from intraoperative MR images was achieved for the first time in gynecological brachytherapy.

Multiatlas based segmentation is largely used in many clinical and research applications. Due to its good performances, it has recently been included in some commercial platforms for radiotherapy planning and surgery guidance. Anyway, to date, a software with no restrictions about the anatomical district and image modality is still missing. In this paper we introduce plastimatch mabs, an open source software that can be used with any image modality for automatic segmentation.plastimatch mabs workflow consists of two main parts: (1) an offline phase, where optimal registration and voting parameters are tuned and (2) an online phase, where a new patient is labeled from scratch by using the same parameters as identified in the former phase. Several registration strategies, as well as different voting criteria can be selected. A flexible atlas selection scheme is also available. To prove the effectiveness of the proposed software across anatomical districts and image modalities, it was tested on two very different scenarios: head and neck (HN) CT segmentation for radiotherapy application, and magnetic resonance image brain labeling for neuroscience investigation.For the neurological study, minimum dice was equal to 0.76 (investigated structures: left and right caudate, putamen, thalamus, and hippocampus). For head and neck case, minimum dice was 0.42 for the most challenging structures (optic nerves and submandibular glands) and 0.62 for the other ones (mandible, brainstem, and parotid glands). Time required to obtain the labels was compatible with a real clinical workflow (35 and 120 min).The proposed software fills a gap in the multiatlas based segmentation field, since all currently available tools (both for commercial and for research purposes) are restricted to a well specified application. Furthermore, it can be adopted as a platform for exploring MABS parameters and as a reference implementation for comparing against other segmentation algorithms.

Purpose: To improve the contouring of clinical target volume for the radiotherapy of neck Hodgkin/non-Hodgkin lymphoma by localizing the prechemotherapy gross target volume onto the simulation computed tomography using [18F]-fluorodeoxyglucose positron emission tomography/computed tomography. Material and Methods: The gross target volume delineated on prechemotherapy [18F]-fluorodeoxyglucose positron emission tomography/computed tomography images was warped onto simulation computed tomography using deformable image registration. Fifteen patients with neck Hodgkin/non-Hodgkin lymphoma were analyzed. Quality of image registration was measured by computing the Dice similarity coefficient on warped organs at risk. Five radiation oncologists visually scored the localization of automatic gross target volume, ranking it from 1 (wrong) to 5 (excellent). Deformable registration was compared to rigid registration by computing the overlap index between the automatic gross target volume and the planned clinical target volume and quantifying the V95 coverage. Results: The Dice similarity coefficient was 0.80 ± 0.07 (median ± quartiles). The physicians’ survey had a median score equal to 4 (good). By comparing the rigid versus deformable registration, the overlap index increased from a factor of about 4 and the V95 (percentage of volume receiving the 95% of the prescribed dose) went from 0.84 ± 0.38 to 0.99 ± 0.10 (median ± quartiles). Conclusion: This study demonstrates the impact of using deformable registration between prechemotherapy [18F]-fluorodeoxyglucose positron emission tomography/computed tomography and simulation computed tomography, in order to automatically localize the gross target volume for radiotherapy treatment of patients with Hodgkin/non-Hodgkin lymphoma.

Automated segmentation is a frequently applied task in the course of medical imaging. Furthermore, it is a substantial component of image-guided radiotherapy. Atlas based segmentation is one of the most frequently used approach for automated segmentation. Especially for multi-atlas based segmentation, segmentation quality and speed largely depends on the underlying registration and atlas selection strategy. In this work an atlas selection strategy that is based on the correlation of inter-atlas similarities within a set of atlas images is presented. Segmentation quality is analyzed by calculating dice coefficients and 95% Hausdorff distances for the left and right parotid with respect to different numbers of atlases. Results are compared to other state of the art atlas selection strategies. It can be shown that the developed atlas selection technique performs slightly better than NMI-based selection if a low number of atlases is used.

Accurate delineation of organs at risk (OARs) is a precondition for intensity modulated radiation therapy. However, manual delineation of OARs is time consuming and prone to high interobserver variability. Because of image artifacts and low image contrast between different structures, however, the number of available approaches for autosegmentation of structures in the head-neck area is still rather low. In this project, a new approach for automated segmentation of head-neck CT images that combine the robustness of multiatlas-based segmentation with the flexibility of geodesic active contours and the prior knowledge provided by statistical appearance models is presented.The presented approach is using an atlas-based segmentation approach in combination with label fusion in order to initialize a segmentation pipeline that is based on using statistical appearance models and geodesic active contours. An anatomically correct approximation of the segmentation result provided by atlas-based segmentation acts as a starting point for an iterative refinement of this approximation. The final segmentation result is based on using model to image registration and geodesic active contours, which are mutually influencing each other.18 CT images in combination with manually segmented labels of parotid glands and brainstem were used in a leave-one-out cross validation scheme in order to evaluate the presented approach. For this purpose, 50 different statistical appearance models have been created and used for segmentation. Dice coefficient (DC), mean absolute distance and max. Hausdorff distance between the autosegmentation results and expert segmentations were calculated. An average Dice coefficient of DC = 0.81 (right parotid gland), DC = 0.84 (left parotid gland), and DC = 0.86 (brainstem) could be achieved.The presented framework provides accurate segmentation results for three important structures in the head neck area. Compared to a segmentation approach based on using multiple atlases in combination with label fusion, the proposed hybrid approach provided more accurate results within a clinically acceptable amount of time.

Purpose: In this work we present the validation of Plastimatch MABS, an open source software for multi atlas based segmentation of medical images. Methods: The validation was performed on two different clinical datasets: 1) 25 CT image volumes of patients treated for HN 2) 20 MRI series of patients having a neurological diagnosis. For the first set, 8 organs at risk to be spared during the therapy were segmented: mandible, brainstem and both left/right optic nerves, parotid and submandibular glands. For the neurological set, 4 paired brain structures were segmented: left/right caudate, putamen, thalamus and hippocampus. For both cases a leave one out approach was used for validation. Tests were performed on common hardware (Intel Xeon CPU @ 2.5 GHz, 8 GB of RAM). Results: Segmentation accuracy was quantified in terms of Dice Similarity Coefficient (DSC) and 95th percentile of Hausdorff Distance (HD). The average DSCs computed over all the labels were 0.70 for H&N application and 0.85 for neuroscience case. HD values confirm the DSCs (4.92 mm and 1.41 mm). Average segmentation time for the two applications was 35 and 120 mins respectively. Conclusion: Plastimatch MABS has been validated for multi atlas based segmentation of medical images. Accuracy was demonstrated for different anatomical districts and image modalities. The results show that Plastimatch MABS can be used for a wide range of applications. The presented approach can provide fully automatic and robust segmentation and is also compatible with timing requirements for clinical applications.