Search

The current study aimed to predict lymphovascular invasion (LVI) using multiple machine learning algorithms and multi-segmentation positron emission tomography (PET) radiomics in non-small cell lung cancer (NSCLC) patients, offering new avenues for personalized treatment strategies and improving patient outcomes. One hundred and twenty-six patients with NSCLC were enrolled in this study. Various automated and semi-automated PET image segmentation methods were applied, including Local Active Contour (LAC), Fuzzy-C-mean (FCM), K-means (KM), Watershed, Region Growing (RG), and Iterative thresholding (IT) with different percentages of the threshold. One hundred five radiomic features were extracted from each region of interest (ROI). Multiple feature selection methods, including Minimum Redundancy Maximum Relevance (MRMR), Recursive Feature Elimination (RFE), and Boruta, and multiple classifiers, including Multilayer Perceptron (MLP), Logistic Regression (LR), XGBoost (XGB), Naive Bayes (NB), and Random Forest (RF), were employed. Synthetic Minority Oversampling Technique (SMOTE) was also used to determine if it boosts the area under the ROC curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Our results indicated that the combination of SMOTE, IT (with 45% threshold), RFE feature selection and LR classifier showed the best performance (AUC = 0.93, ACC = 0.84, SEN = 0.85, SPE = 0.84) followed by SMOTE, FCM segmentation, MRMR feature selection, and LR classifier (AUC = 0.92, ACC = 0.87, SEN = 1, SPE = 0.84). The highest ACC belonged to the IT segmentation (with 45 and 50% thresholds) alongside Boruta feature selection and the NB classifier without SMOTE (ACC = 0.9, AUC = 0.78 and 0.76, SEN = 0.7, and SPE = 0.94, respectively). Our results indicate that selection of appropriate segmentation method and machine learning algorithm may be helpful in successful prediction of LVI in patients with NSCLC with high accuracy using PET radiomics analysis., Competing Interests: Declarations. Ethics approval: This retrospective study was approved by the Institutional Review Board (IRB) of Tehran University of Medical Sciences (approval ID IR.TUMS.MEDICINE.REC.1397.733). Due to the retrospective nature of the study, the requirement for obtaining written informed consent from patients was waived by the IRB. Conflict of interest: The authors have no conflicts to disclose., (© 2024. The Author(s).)

Purpose: The isocitrate dehydrogenase ( IDH ) mutation has become one of the most important prognostic biomarkers in glioma management, indicating better treatment response and prognosis. IDH mutations confer neomorphic activity leading to the conversion of alpha-ketoglutarate (α-KG) to 2-hydroxyglutarate (2HG). The purpose of this study was to investigate the clinical potential of proton MR spectroscopy ( 1 H-MRS) in identifying IDH -mutant gliomas by detecting characteristic resonances of 2HG and its complex interplay with other clinically relevant metabolites., Materials and Methods: Thirty-two patients with suspected infiltrative glioma underwent a single-voxel (SVS, n = 17) and/or single-slice-multivoxel ( 1 H-MRSI, n = 15) proton MR spectroscopy ( 1 H-MRS) sequence with an optimized echo-time (97 ms) on 3T-MRI. Spectroscopy data were analyzed using the linear combination (LC) model. Cramér-Rao lower bound (CRLB) values of <40% were considered acceptable for detecting 2HG and <20% for other metabolites. Immunohistochemical analyses for determining IDH mutational status were subsequently performed from resected tumor specimens and findings were compared with the results from spectral data. Mann-Whitney and chi-squared tests were performed to ascertain differences in metabolite levels between IDH -mutant and IDH -wild-type gliomas. Receiver operating characteristic (ROC) curve analyses were also performed., Results: Data from eight cases were excluded due to poor spectral quality or non-tumor-related etiology, and final data analyses were performed from 24 cases. Of these cases, 9/12 (75%) were correctly identified as IDH -mutant or IDH -wildtype gliomas through SVS and 10/12 (83%) through 1 H-MRSI with an overall concordance rate of 79% (19/24). The sensitivity, specificity, positive predictive value, and negative predictive value were 80%, 77%, 86%, and 70%, respectively. The metabolite 2HG was found to be significant in predicting IDH -mutant gliomas through the chi-squared test ( p < 0.01). The IDH -mutant gliomas also had a significantly higher NAA/Cr ratio (1.20 ± 0.09 vs. 0.75 ± 0.12 p = 0.016) and lower Glx/Cr ratio (0.86 ± 0.078 vs. 1.88 ± 0.66; p = 0.029) than those with IDH wild-type gliomas. The areas under the ROC curves for NAA/Cr and Glx/Cr were 0.808 and 0.786, respectively., Conclusions: Noninvasive optimized 1 H-MRS may be useful in predicting IDH mutational status and 2HG may serve as a valuable diagnostic and prognostic biomarker in patients with gliomas.

Placenta Accreta Spectrum (PAS) refers to the range of abnormally adhesive and penetrative placental tissue at a uterine scar. PAS is divided into accreta, increta, and percreta based on degree of myometrial invasion. Its incidence has increased, and PAS is now the leading indication for emergency peripartum hysterectomy in the setting of catastrophic hemorrhage from a non-separating placenta. The recent release of the International Federation of Gynecology and Obstetrics (FIGO) guidelines in 2018 coupled with the joint consensus statement from the Society of Abdominal Radiology (SAR) and European Society of Urogenital Radiology (ESUR) in 2020 reflect decades worth of diagnostic and therapeutic advances in this field. Although the increasing role of MRI in PAS diagnosis is evident, the literature on PAS reveals several disparate but conceptually overlapping MRI signs. Identifying and differentiating between placenta increta and percreta on imaging may be quite challenging even with MRI and sometimes even on final pathology. In this review, we aim to (i) provide a clarified understanding of PAS pathophysiology, (ii) comprehensively review and classify MRI signs based on pathophysiologic underpinnings, (iii) highlight shortcomings in the current PAS literature; and (iv) highlight best practice guidelines for imaging diagnosis of PAS., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Purpose: To assess the safety and efficacy of transarterial chemoembolization using a 75-μm drug-eluting embolic (DEE) in patients with unresectable hepatocellular carcinoma (HCC)., Materials and Methods: The medical records of 109 patients with a mean age of 64.1 years (range 85-49) treated for unresectable HCC between November 2013 and August 2016 with transarterial chemoembolization using a 75-μm DEE were retrospectively reviewed. Patients who had prior therapy for HCC were excluded. Child-Pugh A patients and Barcelona Clinic Liver Cancer stages A/B patients constituted 68.8% and 65.1% of the patients, respectively. The mean size of the index tumors was 5.8 cm (range 18.5-1.2) with 42 (39%) patients with central tumors around the porta-hepatis region. Portal vein invasion was seen in 10 (9.2%) patients. Tumor response was categorized according to the modified Response Evaluation Criteria in Solid Tumors 1.1, and the toxicity profile was assessed using Common Terminology Criteria for Adverse Events, version 4.03., Results: At 1-month follow-up, complete response, objective response, and disease control was seen in 23%, 66%, and 90%, respectively. The median progression-free survival was 11.2 months. The median overall survival was 25.1 months (33.4 months for Child-Pugh A and 28.2 months for Barcelona Clinic Liver Cancer stages A/B), and transplant-free survival was 21.3 months. The 6-, 12-, and 24-month survivals were 91.7%, 75.5%, and 50.5%, respectively. Grade 3 toxicity was seen in 1.8% of the patients; no grade 4 or 5 toxicity was reported., Conclusions: Transarterial chemoembolization using 75-μm DEE is safe and efficacious in the treatment of HCC., (Copyright © 2019 SIR. Published by Elsevier Inc. All rights reserved.)

Purpose: To test the hypothesis that ruptured abdominal aortic aneurysms (AAA) are globally weaker than unruptured ones., Methods: Four ruptured and seven unruptured AAA specimens were harvested whole from fresh cadavers during autopsies performed over an 18-month period. Multiple regionally distributed longitudinally oriented rectangular strips were cut from each AAA specimen for a total of 77 specimen strips. Strips were subjected to uniaxial extension until failure. Sections from approximately the strongest and weakest specimen strips were studied histologically and histochemically. From the load-extension data, failure tension, failure stress and failure strain were calculated. Rupture site characteristics such as location, arc length of rupture and orientation of rupture were also documented., Results: The failure tension, a measure of the tissue mechanical caliber was remarkably similar between ruptured and unruptured AAA (group mean ± standard deviation of within-subject means: 11.2±2.3 versus 11.6±3.6N/cm; p=0.866 by mixed model ANOVA). In post-hoc analysis, there was little difference between the groups in other measures of tissue mechanical caliber as well such as failure stress (95±28 versus 98±23 N/cm(2); p=0.870), failure strain (0.39±0.09 versus 0.36±0.09; p=0.705), wall thickness (1.7±0.4 versus 1.5±0.4mm; p=0.470) , and % coverage of collagen within tissue cross section (49.6±12.9% versus 60.8±9.6%; p=0.133). In the four ruptured AAA, primary rupture sites were on the lateral quadrants (two on left; one on left-posterior; one on right). Remarkably, all rupture lines had a longitudinal orientation and ranged from 1 to 6 cm in length., Conclusion: The findings are not consistent with the hypothesis that ruptured aortic aneurysms are globally weaker than unruptured ones., (Copyright © 2011 Elsevier Ltd. All rights reserved.)

The regional distribution of wall thickness and failure properties in human abdominal aortic aneurysm (AAA) was explored. Three unruptured and one ruptured AAA were harvested as a whole during necropsy. Thickness was measured at about every 1.5 cm(2) wall surface area for an average of 100 measurement sites per AAA. Multiple longitudinally oriented rectangular specimen strips were cut at various locations from each AAA for a total of 48 strips. The strips were subjected to uniaxial extension until failure. Wall thickness varied regionally and between AAA from as low as 0.23 mm at a rupture site to 4.26 mm at a calcified site (median=1.48 mm). Wall thickness was slightly lower in the posterior and right regions. The failure tension (ultimate) of specimen strips varied regionally and between AAA from 5.5 N/cm close to a blister site in the ruptured AAA to 42.3N/cm at the undilated neck of a 4 cm diameter unruptured AAA (median=14.8 N/cm). Failure stress (ultimate) varied from 33.6 to 235.1N/cm(2) (median=126.6N/cm(2)). There was no perceptible pattern in failure properties along the circumference. Failure tension of specimen strips at or close to blisters was mostly low. The rupture site in the ruptured aneurysm had the lowest recorded wall thickness of 0.23 mm with only slightly higher readings within a 1cm radius. The failure tension of the specimen strip close to the rupture site was low (11.1 N/cm) compared to its neighborhood in the ruptured aneurysm.