Evaluating the quality of radiomics-based studies for endometrial cancer using RQS and METRICS tools.

OBJECTIVE: To assess the methodological quality of radiomics-based models in endometrial cancer using the radiomics quality score (RQS) and METhodological radiomICs score (METRICS). METHODS: We systematically reviewed studies published by October 30th, 2023. Inclusion criteria were original radiomics studies on endometrial cancer using CT, MRI, PET, or ultrasound. Articles underwent a quality assessment by novice and expert radiologists using RQS and METRICS. The inter-rater reliability for RQS and METRICS among radiologists with varying expertise was determined. Subgroup analyses were performed to assess whether scores varied according to study topic, imaging technique, publication year, and journal quartile. RESULTS: Sixty-eight studies were analysed, with a median RQS of 11 (IQR, 9-14) and METRICS score of 67.6% (IQR, 58.8-76.0); two different articles reached maximum RQS of 19 and METRICS of 90.7%, respectively. Most studies utilised MRI (82.3%) and machine learning methods (88.2%). Characterisation and recurrence risk stratification were the most explored outcomes, featured in 35.3% and 19.1% of articles, respectively. High inter-rater reliability was observed for both RQS (ICC: 0.897; 95% CI: 0.821, 0.946) and METRICS (ICC: 0.959; 95% CI: 0.928, 0.979). Methodological limitations such as lack of external validation suggest areas for improvement. At subgroup analyses, no statistically significant difference was noted. CONCLUSIONS: Whilst using RQS, the quality of endometrial cancer radiomics research was apparently unsatisfactory, METRICS depicts a good overall quality. Our study highlights the need for strict compliance with quality metrics. Adhering to these quality measures can increase the consistency of radiomics towards clinical application in the pre-operative management of endometrial cancer. CLINICAL RELEVANCE STATEMENT: Both the RQS and METRICS can function as instrumental tools for identifying different methodological deficiencies in endometrial cancer radiomics research. However, METRICS also reflected a focus on the practical applicability and clarity of documentation. KEY POINTS: The topic of radiomics currently lacks standardisation, limiting clinical implementation. METRICS scores were generally higher than the RQS, reflecting differences in the development process and methodological content. A positive trend in METRICS score may suggest growing attention to methodological aspects in radiomics research.

Distinguishing Pituitary Metastasis and Pituitary Neuroendocrine Tumors through Conventional MR Imaging and Clinical Features.

BACKGROUND AND PURPOSE: Given their overlapping features, pituitary metastases frequently imitate pituitary neuroendocrine tumors in neuroimaging studies. This study aimed to distinguish pituitary metastases from pituitary neuroendocrine tumors on the basis of conventional MR imaging and clinical features as a practical approach. MATERIALS AND METHODS: In this 2-center retrospective study, backward from January 2024, preoperative pituitary MR imaging examinations of 22 pituitary metastases and 74 pituitary neuroendocrine tumors were analyzed. Exclusion criteria were as follows: absence of a definitive histopathologic diagnosis, history of pituitary surgery or radiation therapy before MR imaging, and pituitary neuroendocrine tumors treated with medical therapy. Two radiologists systematically evaluated 13 conventional MR imaging features that have been reported more commonly as indicative of pituitary metastases and pituitary neuroendocrine tumors in the literature. Age, sex, history of cancer, and maximum tumor size constituted the clinical/epidemiologic features. The primary cancer origin for this study was also noted. Univariable and multivariable logistic regression was used for the selection of variables, determining independent predictors, and modeling. Interobserver agreement was evaluated for all imaging parameters using the Cohen κ statistic or intraclass correlation coefficient. RESULTS: A total of 22 patients with pituitary metastases (8 women; mean age, 49.5 [SD, 13] years) and 74 patients with pituitary neuroendocrine tumors (36 women; mean age, 50.1 [SD, 11] years) were enrolled. There was no statistically significant distributional difference in age, sex, or maximum tumor size between the 2 groups. Lung cancer (9/22; 41%) was the most commonly reported primary tumor, followed by breast (3/22; 13.6%) and unknown cancer (3/22; 13.6%). Logistic regression revealed 3 independent predictors: rapid growth on control MR imaging, masslike or nodular expansion of the pituitary stalk, and a history of cancer. The model based on these 3 features achieved an area under the curve, accuracy, sensitivity, specificity, and Brier score of 0.987 (95% CI, 0.964-1), 97.9% (95% CI, 92.7%-99.8%), 95.5% (95% CI, 77.2%-99.9%), 98.6% (95% CI, 92.7%-100%), and 0.025, respectively. CONCLUSIONS: Two conventional features based on pituitary MR imaging with the clinical variable of history of cancer had satisfying predictive performance, making them potential discriminators between pituitary metastases and pituitary neuroendocrine tumors. In cases in which differentiation between pituitary metastases and pituitary neuroendocrine tumors poses a challenge, the results of this study may help with the diagnosis.

Utility of diffusion tensor imaging and generalized q-sampling imaging for predicting short-term clinical effect of deep brain stimulation in Parkinson's disease.

PURPOSE: To assess whether diffusion tensor imaging (DTI) and generalized q-sampling imaging (GQI) metrics could preoperatively predict the clinical outcome of deep brain stimulation (DBS) in patients with Parkinson's disease (PD). METHODS: In this single-center retrospective study, from September 2021 to March 2023, preoperative DTI and GQI examinations of 44 patients who underwent DBS surgery, were analyzed. To evaluate motor functions, the Unified Parkinson's Disease Rating Scale (UPDRS) during on- and off-medication and Parkinson's Disease Questionnaire-39 (PDQ-39) scales were used before and three months after DBS surgery. The study population was divided into two groups according to the improvement rate of scales: ≥ 50% and < 50%. Five target regions, reported to be affected in PD, were investigated. The parameters having statistically significant difference were subjected to a receiver operating characteristic (ROC) analysis. RESULTS: Quantitative anisotropy (qa) values from globus pallidus externus, globus pallidus internus (qa_Gpi), and substantia nigra exhibited significant distributional difference between groups in terms of the improvement rate of UPDRS-3 scale during on-medication (p = 0.003, p = 0.0003, and p = 0.0008, respectively). In ROC analysis, the best parameter in predicting DBS response included qa_Gpi with a cut-off value of 0.01370 achieved an area under the ROC curve, accuracy, sensitivity, and specificity of 0.810, 73%, 62.5%, and 85%, respectively. Optimal cut-off values of ≥ 0.01864 and ≤ 0.01162 yielded a sensitivity and specificity of 100%, respectively. CONCLUSION: The imaging parameters acquired from GQI, particularly qa_Gpi, may have the ability to non-invasively predict the clinical outcome of DBS surgery.

Effects of erector spinae plane block on opioid consumption in patients undergoing hand-assisted laparoscopic donor nephrectomy: a randomized controlled trial.

BACKGROUND: The erector spinae plane block is a relatively new regional anesthesia technique that is expected to provide some benefits for postoperative analgesia. This study investigated the effects of erector spinae plane block on postoperative opioid consumption in kidney donors undergoing hand-assisted laparoscopic donor nephrectomy for renal transplantation. METHODS: Fifty-two donors scheduled for elective hand-assisted laparoscopic donor nephrectomy were randomly divided into the block (25 donors) and control (27 donors) groups. Donors in the block group received 30 mL of 0.25% bupivacaine under ultrasound guidance, whereas the control group received no block treatment. The primary outcome measure was the amount of fentanyl administered via patient-controlled analgesia at 24 h. Secondary outcomes included the duration of stay, opioid consumption in the post-anesthesia care unit, and pain scores during the recording hours. RESULTS: No significant differences were observed between the groups regarding total opioid consumption converted to intravenous morphine equivalent administered via patient-controlled analgesia (33.3±21.4 mg vs. 37.5±18.5 mg; P=0.27) and in the postanesthesia care unit (1.5±0.9 mg vs. 1.4±0.8 mg; P=0.55). The duration of stay in the postanesthesia care unit (86.3±32.6 min vs. 85.7±33.6 min; P=0.87) was similar between the groups. There was no significant difference between the groups in the postoperative donor-reported NRS pain scores (P>0.05 for all the time points). CONCLUSIONS: Preoperative erector spinae plane block is not an effective strategy for reducing postoperative pain or opioid consumption in patients undergoing hand-assisted laparoscopic donor nephrectomy. Different block combinations are needed for optimal pain management in hand-assisted laparoscopic donor nephrectomy.

Reproducibility of rCBV in glioblastomas using T2*-weighted perfusion MRI: an evaluation of sampling, normalization, and experience.

PURPOSE: The reproducibility of relative cerebral blood volume (rCBV) measurements among readers with different levels of experience is a concern. This study aimed to investigate the inter-reader reproducibility of rCBV measurement of glioblastomas using the hotspot method in dynamic susceptibility contrast perfusion magnetic resonance imaging (DSC-MRI) with various strategies. METHODS: In this institutional review board-approved single-center study, 30 patients with glioblastoma were retrospectively evaluated with DSC-MRI at a 3.0 Tesla scanner. Three groups of reviewers, including neuroradiologists, general radiologists, and radiology residents, calculated the rCBV based on the number of regions of interest (ROIs) and reference areas. For statistical analysis of feature reproducibility, the intraclass correlation coefficient (ICC) and Bland-Altman plots were used. Analyses were made among individuals, reader groups, reader-group pooling, and a population that contained all of them. RESULTS: For individuals, the highest inter-reader reproducibility was observed between neuroradiologists [ICC: 0.527; 95% confidence interval (CI): 0.21-0.74] and between residents (ICC: 0.513; 95% CI: 0.20-0.73). There was poor reproducibility in the analyses of individuals with different levels of experience (ICC range: 0.296-0.335) and in reader-wise and group-wise pooling (ICC range: 0.296-0.335 and 0.397-0.427, respectively). However, an increase in ICC values was observed when five ROIs were used. In an analysis of all strategies, the ICC for the centrum semiovale was significantly higher than that for contralateral white matter (P < 0.001). CONCLUSION: The inter-reader reproducibility of rCBV measurement was poor to moderate regardless of whether it was calculated by neuroradiologists, general radiologists, or residents, which may indicate the need for automated methods. Choosing five ROIs and using the centrum semiovale as a reference area may increase reliability for all users.

Influence of image preprocessing on the segmentation-based reproducibility of radiomic features: in vivo experiments on discretization and resampling parameters.

PURPOSE: To systematically investigate the impact of image preprocessing parameters on the segmentation-based reproducibility of magnetic resonance imaging (MRI) radiomic features. METHODS: The MRI scans of 50 patients were included from the multi-institutional Brain Tumor Segmentation 2021 public glioma dataset. Whole tumor volumes were manually segmented by two independent readers, with the participation of eight readers. Radiomic features were extracted from two sequences: T2-weighted (T2) and contrast-enhanced T1-weighted (T1ce). Two methods were considered for discretization: bin count (i.e., relative discretization) and bin width (i.e., absolute discretization). Ten discretization (five for each method) and five resampling parameters were varied while other parameters were fixed. The intraclass correlation coefficient (ICC) was used for reliability analysis based on two commonly used cut-off values (0.75 and 0.90). RESULTS: Image preprocessing parameters had a significant impact on the segmentation-based reproducibility of radiomic features. The bin width method yielded more reproducible features than the bin count method. In discretization experiments using the bin width on both sequences, according to the ICC cut-off values of 0.75 and 0.90, the rate of reproducible features ranged from 70% to 84% and from 35% to 57%, respectively, with an increasing percentage trend as parameter values decreased (from 84 to 5 for T2; 100 to 6 for T1ce). In the resampling experiments, these ranged from 53% to 74% and from 10% to 20%, respectively, with an increasing percentage trend from lower to higher parameter values (physical voxel size; from 1 x 1 x 1 to 2 x 2 x 2 mm3). CONCLUSION: The segmentation-based reproducibility of radiomic features appears to be substantially influenced by discretization and resampling parameters. Our findings indicate that the bin width method should be used for discretization and lower bin width and higher resampling values should be used to allow more reproducible features.

Predicting Isocitrate Dehydrogenase Mutation Status of Grade 2-4 Gliomas with Diffusion Tensor Imaging (DTI) Parameters Derived from Model-Based DTI and Model-Free Q-Sampling Imaging Reconstructions.

OBJECTIVE: To determine whether diffusion tensor imaging (DTI) parameters acquired with model-based DTI and model-free generalized Q-sampling imaging (GQI) reconstructions may noninvasively predict isocitrate dehydrogenase (IDH) mutational status in patients with grade 2-4 gliomas. METHODS: Forty patients with known IDH genotype (28 IDH wild-type; 12 IDH mutant) who underwent preoperative DTI evaluation on a 3-Tesla magnetic resonance imaging scanner were analyzed retrospectively. Absolute values obtained from model-based and model-free reconstructions were compared. Using the intraclass correlation coefficient, interobserver agreement was assessed for various sampling techniques. Variables having statistically significant distributions between IDH groups were subjected to a receiver operating characteristic (ROC) analysis. Using multivariable logistic regression analysis, independent predictors, if present, were identified and a model was developed. RESULTS: Six imaging parameters (3 from model-based DTI and 3 from model-free GQI reconstructions) showed statistically significant differences between groups (P < 0.001, power >0.97), with very high correlation to each other (P < 0.001). Age difference between the groups was statistically significant (P < 0.001). The optimal logistic regression model comprised a GQI-based parameter and age, which were independent predictors as well, producing an area under the ROC curve, accuracy, sensitivity, and specificity of 0.926, 85%, 75%, and 89.3%, respectively. Using the GQI reconstruction feature alone with a cut-off of 1.60, an 85% of accuracy was also achieved with ROC analysis. CONCLUSIONS: The imaging parameters acquired from model-based DTI and model-free GQI reconstructions, combined with the clinical variable age, may have the ability to noninvasively predict the IDH genotype in gliomas, either alone or in particular combinations.

Experience of kidney transplantation to a patient with Bernard Soulier syndrome: A case report.

Renal transplantation could be a challenging operation in patients with haemorrhagic diathesis, with predictable difficulties or even with unpredictable hurdles. Bernard Soulier Syndrome (BSS) is one of the ethiologies of the thrombocytopenia and it is a rare hereditary disease associated with defects of the platelet glycoprotein complex glycoprotein Ib/V/IX and characterized by large platelets, thrombocytopenia, and severe bleeding symptoms. Here, we present a challenging renal transplantation in BSS.

REliability of consensus-based segMentatIoN in raDiomic feature reproducibility (REMIND): A word of caution.

OBJECTIVE: To evaluate the reliability of consensus-based segmentation in terms of reproducibility of radiomic features. METHODS: In this retrospective study, three tumor data sets were investigated: breast cancer (n = 30), renal cell carcinoma (n = 30), and pituitary macroadenoma (n = 30). MRI was utilized for breast and pituitary data sets, while CT was used for renal data set. 12 readers participated in the segmentation process. Consensus segmentation was created by making corrections on a previous region or volume of interest. Four experiments were designed to evaluate the reproducibility of radiomic features. Reliability was assessed with intraclass correlation coefficient (ICC) with two cut-off values: 0.75 and 0.9. RESULTS: Considering the lower bound of the 95% confidence interval and the ICC threshold of 0.90, at least 61% of the radiomic features were not reproducible in the inter-consensus analysis. In the susceptibility experiment, at least half (54%) became non-reproducible when the first reader is replaced with a different reader. In the intra-consensus analysis, at least about one-third (32%) were non-reproducible when the same second reader segmented the image over the same first reader two weeks later. Compared to inter-reader analysis based on independent single readers, the inter-consensus analysis did not statistically significantly improve the rates of reproducible features in all data sets and analyses. CONCLUSIONS: Despite the positive connotation of the word "consensus", it is essential to REMIND that consensus-based segmentation has significant reproducibility issues. Therefore, the usage of consensus-based segmentation alone should be avoided unless a reliability analysis is performed, even if it is not practical in clinical settings.

Trends and statistics of artificial intelligence and radiomics research in Radiology, Nuclear Medicine, and Medical Imaging: bibliometric analysis.

OBJECTIVE: To conduct a comprehensive bibliometric analysis of artificial intelligence (AI) and its subfields as well as radiomics in Radiology, Nuclear Medicine, and Medical Imaging (RNMMI). METHODS: Web of Science was queried for relevant publications in RNMMI and medicine along with their associated data from 2000 to 2021. Bibliometric techniques utilised were co-occurrence, co-authorship, citation burst, and thematic evolution analyses. Growth rate and doubling time were also estimated using log-linear regression analyses. RESULTS: According to the number of publications, RNMMI (11,209; 19.8%) was the most prominent category in medicine (56,734). USA (44.6%) and China (23.1%) were the two most productive and collaborative countries. USA and Germany experienced the strongest citation bursts. Thematic evolution has recently exhibited a significant shift toward deep learning. In all analyses, the annual number of publications and citations demonstrated exponential growth, with deep learning-based publications exhibiting the most prominent growth pattern. Estimated continuous growth rate, annual growth rate, and doubling time of the AI and machine learning publications in RNMMI were 26.1% (95% confidence interval [CI], 12.0-40.2%), 29.8% (95% CI, 12.7-49.5%), and 2.7 years (95% CI, 1.7-5.8), respectively. In the sensitivity analysis using data from the last 5 and 10 years, these estimates ranged from 47.6 to 51.1%, 61.0 to 66.7%, and 1.4 to 1.5 years. CONCLUSION: This study provides an overview of AI and radiomics research conducted mainly in RNMMI. These results may assist researchers, practitioners, policymakers, and organisations in gaining a better understanding of both the evolution of these fields and the importance of supporting (e.g., financial) these research activities. KEY POINTS: • In terms of the number of publications on AI and ML, Radiology, Nuclear Medicine, and Medical Imaging was the most prominent category compared to the other categories related to medicine (e.g., Health Policy & Services, Surgery). • All evaluated analyses (i.e., AI, its subfields, and radiomics), based on the annual number of publications and citations, demonstrated exponential growth, with decreasing doubling time, which indicates increasing interest from researchers, journals, and, in turn, the medical imaging community. • The most prominent growth pattern was observed in deep learning-based publications. However, the further thematic analysis demonstrated that deep learning has been underdeveloped but highly relevant to the medical imaging community.

Use of Thalamus L-Sign to Differentiate Periventricular Leukomalacia From Neurometabolic Disorders.

PURPOSE: To assess the diagnostic value of the thalamus L-sign on magnetic resonance imaging (MRI) in distinguishing between periventricular leukomalacia and neurometabolic disorders in pediatric patients. METHODS: In this retrospective study, clinical and imaging information was collected from 50 children with periventricular leukomalacia and 52 children with neurometabolic disorders. MRI was used to evaluate the L-sign of the thalamus (ie, injury to the posterolateral thalamus) and the lobar distribution of signal intensity changes. Age, sex, gestational age, and level of Gross Motor Function Classification System (only for periventricular leukomalacia) constituted the clinical parameters. Statistical evaluation of group differences for imaging and clinical variables were conducted using univariable statistical methods. The intra- and inter-observer agreement was evaluated using Cohen's kappa. Univariable or multivariable logistic regression was employed for selection of variables, determining independent predictors, and modeling. RESULTS: The thalamus L-sign was observed in 70% (35/50) of patients in the periventricular leukomalacia group, but in none of the patients with neurometabolic disorder (P < .001). The gestational age between groups varied significantly (P < .001). Involvement of frontal, parietal, and occipital lobes differed significantly between groups (P < .001). In the logistic regression, the best model included negative thalamus L-sign and gestational age, yielding an area under the curve, accuracy, sensitivity, specificity, and precision values of 0.995, 96.1%, 96%, 96.2%, and 96%, respectively. Both the lack of thalamus L-sign and gestational age were independent predictors (P < .001). CONCLUSIONS: The thalamus L-sign and gestational age may be useful in distinguishing between periventricular leukomalacia and neurometabolic disorders.

Radiomics of locally advanced rectal cancer: machine learning-based prediction of response to neoadjuvant chemoradiotherapy using pre-treatment sagittal T2-weighted MRI.

PURPOSE: Variable response to neoadjuvant chemoradiotherapy (nCRT) is observed among individuals with locally advanced rectal cancer (LARC), having a significant impact on patient management. In this work, we aimed to investigate the potential value of machine learning (ML)-based magnetic resonance imaging (MRI) radiomics in predicting therapeutic response to nCRT in patients with LARC. MATERIALS AND METHODS: Seventy-six patients with LARC were included in this retrospective study. Radiomic features were extracted from pre-treatment sagittal T2-weighted MRI images, with 3D segmentation. Dimension reduction was performed with a reliability analysis, pair-wise correlation analysis, analysis of variance, recursive feature elimination, Kruskal-Wallis, and Relief methods. Models were created using four different algorithms. In addition to radiomic models, clinical only and different combined models were developed and compared. The reference standard was tumor regression grade (TRG) based on the Modified Ryan Scheme (TRG 0 vs TRG 1-3). Models were compared based on net reclassification index (NRI). Clinical utility was assessed with decision curve analysis (DCA). RESULTS: Number of features with excellent reliability is 106. The best result was achieved with radiomic only model using eight features. The area under the curve (AUC), accuracy, sensitivity, and specificity for validation were 0.753 (standard deviation [SD], 0.082), 81.1%, 83.8%, and 75.0%; for testing, 0.705 (SD, 0.145), 73.9%, 81.2%, and 57.1%, respectively. Based on the clinical only model as reference, NRI for radiomic only model was the best. DCA also showed better clinical utility for radiomic only model. CONCLUSIONS: ML-based T2-weighted MRI radiomics might have a potential in predicting response to nCRT in patients with LARC.

Hand-assisted laparoscopic donor nephrectomy: 1864 cases in 15 years of experience.

BACKGROUND: To evaluate hand-assisted laparoscopic donor nephrectomy (HALDN) in terms of intraoperative and postoperative results. METHODS: After institutional review board approval was obtained, a total of 1864 HALDN operations performed between March 2007 and January 2022 were retrospectively analyzed. Age, sex, body mass index (BMI), status of smoking and presence of previous abdominal surgery, laterality, operative time, transfusion requirement, port counts, length of extraction incision, time until mobilization, time until oral intake, donor serum creatinine levels before and one week after the surgery, length of postoperative hospital stay, intraoperative complications, and postoperative recovery and complications were recorded and statistically analyzed. Multiple renal arteries, BMI, right nephrectomy and male sex were also separately evaluated as risk factors for complications and operative time. RESULTS: A total of 825 (44.26%) male and 1039 (55.74%) female patients were enrolled in the study. The mean age of the patients was 45.79 ± 12.88 years. There were a total of 143 complications (7.67% of the total 1864 cases) consisting of 68 (3.65%) intraoperative and 75 (4.02%) postoperative complications. Open conversion was necessary for 10 patients (0.53%) to manage intraoperative complications. Reoperation was needed for 1 patient due to bleeding 6 h after the operation. Multiple renal arteries were a risk factor for intraoperative complications and prolonged operative time. Right nephrectomy and male sex were also related with longer operative times. DISCUSSION: HALDN is a safe procedure associated with low complication rates.

Radioproteomics in Breast Cancer: Prediction of Ki-67 Expression With MRI-based Radiomic Models.

RATIONALE AND OBJECTIVES: We aimed to investigate the value of magnetic resonance image (MRI)-based radiomics in predicting Ki-67 expression of breast cancer. METHODS: In this retrospective study, 159 lesions from 154 patients were included. Radiomic features were extracted from contrast-enhanced T1-weighted MRI (C+MRI) and apparent diffusion coefficient (ADC) maps, with open-source software. Dimension reduction was done with reliability analysis, collinearity analysis, and feature selection. Two different Ki-67 expression cut-off values (14% vs 20%) were studied as reference standard for the classifications. Input for the models were radiomic features from individual MRI sequences or their combination. Classifications were performed using a generalized linear model. RESULTS: Considering Ki-67 cut-off value of 14%, training and testing AUC values were 0.785 (standard deviation [SD], 0.193) and 0.849 for ADC; 0.696 (SD, 0.150) and 0.695 for C+MRI; 0.755 (SD, 0.171) and 0.635 for the combination of both sequences, respectively. Regarding Ki-67 cut-off value of 20%, training and testing AUC values were 0.744 (SD, 0.197) and 0.617 for ADC; 0.629 (SD, 0.251) and 0.741 for C+MRI; 0.761 (SD, 0.207) and 0.618 for the combination of both sequences, respectively. CONCLUSION: ADC map-based selected radiomic features coupled with generalized linear modeling might be a promising non-invasive method to determine the Ki-67 expression level of breast cancer.

MRI Radiomics of Breast Cancer: Machine Learning-Based Prediction of Lymphovascular Invasion Status.

RATIONALE AND OBJECTIVES: In patients with breast cancer (BC), lymphovascular invasion (LVI) status is considered an important prognostic factor. We aimed to develop machine learning (ML)-based radiomics models for the prediction of LVI status in patients with BC, using preoperative MRI images. MATERIALS AND METHODS: This retrospective study included patients with BC with known LVI status and preoperative MRI. The dataset was split into training and unseen testing sets by stratified sampling with a 2:1 ratio. 2D and 3D radiomic features were extracted from contrast-enhanced T1 weighted images (C+T1W) and apparent diffusion coefficient (ADC) maps. The reliability of the features was assessed with two radiologists' segmentation data. Dimension reduction was done with reliability analysis, multi-collinearity analysis, removal of low-variance features, and feature selection. ML models were created with base, tuned, and boosted random forest algorithms. RESULT: A total of 128 lesions (LVI-positive, 76; LVI-negative, 52) were included. The best model performance was achieved with tunning and boosting model based on 3D ADC maps and selected four radiomic features. The area under the curve and accuracy were 0.726 and 63.5% in the training data, 0.732 and 76.7% in the test data, respectively. The overall sensitivity and positive predictive values were 68% and 69.6% in the training data, 84.6% and 78.6% in the test data, respectively. CONCLUSION: ML and radiomics based on 3D segmentation of ADC maps can be used to predict LVI status in BC, with satisfying performance.

Machine Learning-Based Multiparametric Magnetic Resonance Imaging Radiomics for Prediction of H3K27M Mutation in Midline Gliomas.

OBJECTIVE: H3K27M mutation in gliomas has prognostic implications. Previous magnetic resonance imaging (MRI) studies have reported variable rates of tumoral enhancement, necrotic changes, and peritumoral edema in H3K27M-mutant gliomas, with no distinguishing imaging features compared with wild-type gliomas. We aimed to construct an MRI machine learning (ML)-based radiomic model to predict H3K27M mutation in midline gliomas. METHODS: A total of 109 patients from 3 academic centers were included in this study. Fifty patients had H3K27M mutation and 59 were wild-type. Conventional MRI sequences (T1-weighted, T2-weighted, T2-fluid-attenuated inversion recovery, postcontrast T1-weighted, and apparent diffusion coefficient maps) were used for feature extraction. A total of 651 radiomic features per each sequence were extracted. Patients were randomly selected with a 7:3 ratio to create training (n = 76) and test (n = 33) data sets. An extreme gradient boosting algorithm (XGBoost) was used in ML-based model development. Performance of the model was assessed by area under the receiver operating characteristic curve. RESULTS: Pediatric patients accounted for a larger proportion of the study cohort (60 pediatric [55%] vs. 49 adult [45%] patients). XGBoost with additional feature selection had an area under the receiver operating characteristic curve of 0.791 and 0.737 in the training and test data sets, respectively. The model achieved accuracy, precision (positive predictive value), recall (sensitivity), and F1 (harmonic mean of precision and recall) measures of 72.7%, 76.5%, 72.2%, and 74.3%, respectively, in the test set. CONCLUSIONS: Our multi-institutional study suggests that ML-based radiomic analysis of multiparametric MRI can be a promising noninvasive technique to predict H3K27M mutation status in midline gliomas.

Extreme living donation: A single center simultaneous and sequential living liver-kidney donor experience with long-term outcomes under literature review.

OBJECTIVES: Living liver and kidney donor surgeries are major surgical procedures applied to healthy people with mortality and morbidity risks not providing any direct therapeutic advantage to the donor. In this study, we aimed to share our simultaneous and sequential living liver-kidney donor experience under literature review in this worldwide rare practice. MATERIAL AND METHODS: Between January 2007 and February 2018, a total of 1109 living donor nephrectomies and 867 living liver donor hepatectomies were performed with no mortality to living-related donors. Eight donors who were simultaneous or sequential living liver-kidney donors in this time period were retrospectively reviewed and presented with their minimum 2- year follow-up. RESULTS: Of the 8 donors, 3 of them were simultaneous and 5 of them were sequential liver-kidney donation. All of them were close relatives. Mean age was 39 (26-61) years and mean BMI was 25.7 (17.7-40). In 3 donors, right lobe, in 4 donors, left lateral sector, and in 1 donor, left lobe hepatectomy were performed. Median hospital stay was 9 (7-13) days. Two donors experienced early and late postoperative complications (Grade 3b and Grade 1). No mortality and no other long-term complication occurred. CONCLUSION: Expansion of the donor pool by utilizing grafts from living donors is a globally-accepted proposition since it provides safety and successful outcomes. Simultaneous or sequential liver and kidney donation from the same donor seems to be a reasonable option for combined liver-kidney transplant recipients in special circumstances with acceptable outcomes.

Tubular gastric adenocarcinoma: machine learning-based CT texture analysis for predicting lymphovascular and perineural invasion.

PURPOSE: Lymphovascular invasion (LVI) and perineural invasion (PNI) are associated with poor prognosis in gastric cancers. In this work, we aimed to investigate the potential role of computed tomography (CT) texture analysis in predicting LVI and PNI in patients with tubular gastric adenocarcinoma (GAC) using a machine learning (ML) approach. METHODS: Sixty-eight patients who underwent total gastrectomy with curative (R0) resection and D2-lymphadenectomy were included in this retrospective study. Texture features were extracted from the portal venous phase CT images. Dimension reduction was first done with a reproducibility analysis by two radiologists. Then, a feature selection algorithm was used to further reduce the high-dimensionality of the radiomic data. Training and test splits were created with 100 random samplings. ML-based classifications were done using adaptive boosting, k-nearest neighbors, Naive Bayes, neural network, random forest, stochastic gradient descent, support vector machine, and decision tree. Predictive performance of the ML algorithms was mainly evaluated using the mean area under the curve (AUC) metric. RESULTS: Among 271 texture features, 150 features had excellent reproducibility, which were included in the further feature selection process. Dimension reduction steps yielded five texture features for LVI and five for PNI. Considering all eight ML algorithms, mean AUC and accuracy ranges for predicting LVI were 0.777-0.894 and 76%-81.5%, respectively. For predicting PNI, mean AUC and accuracy ranges were 0.482-0.754 and 54%-68.2%, respectively. The best performances for predicting LVI and PNI were achieved with the random forest and Naive Bayes algorithms, respectively. CONCLUSION: ML-based CT texture analysis has a potential for predicting LVI and PNI of the tubular GACs. Overall, the method was more successful in predicting LVI than PNI.

Artificial Intelligence in Renal Mass Characterization: A Systematic Review of Methodologic Items Related to Modeling, Performance Evaluation, Clinical Utility, and Transparency.

OBJECTIVE. The objective of our study was to systematically review the literature about the application of artificial intelligence (AI) to renal mass characterization with a focus on the methodologic quality items. MATERIALS AND METHODS. A systematic literature search was conducted using PubMed to identify original research studies about the application of AI to renal mass characterization. Besides baseline study characteristics, a total of 15 methodologic quality items were extracted and evaluated on the basis of the following four main categories: modeling, performance evaluation, clinical utility, and transparency items. The qualitative synthesis was presented using descriptive statistics with an accompanying narrative. RESULTS. Thirty studies were included in this systematic review. Overall, the methodologic quality items were mostly favorable for modeling (63%) and performance evaluation (63%). Even so, the studies (57%) more frequently constructed their work on nonrobust features. Furthermore, only a few studies (10%) had a generalizability assessment with independent or external validation. The studies were mostly unsuccessful in terms of clinical utility evaluation (89%) and transparency (97%) items. For clinical utility, the interesting findings were lack of comparisons with both radiologists' evaluation (87%) and traditional models (70%) in most of the studies. For transparency, most studies (97%) did not share their data with the public. CONCLUSION. To bring AI-based renal mass characterization from research to practice, future studies need to improve modeling and performance evaluation strategies and pay attention to clinical utility and transparency issues.