Petrović, Bojan

The translation of AI-generated brain metastases (BM) segmentation into clinical practice relies heavily on diverse, high-quality annotated medical imaging datasets. The BraTS-METS 2023 challenge has gained momentum for testing and benchmarking algorithms using rigorously annotated internationally compiled real-world datasets. This study presents the results of the segmentation challenge and characterizes the challenging cases that impacted the performance of the winning algorithms. Untreated brain metastases on standard anatomic MRI sequences (T1, T2, FLAIR, T1PG) from eight contributed international datasets were annotated in stepwise method: published UNET algorithms, student, neuroradiologist, final approver neuroradiologist. Segmentations were ranked based on lesion-wise Dice and Hausdorff distance (HD95) scores. False positives (FP) and false negatives (FN) were rigorously penalized, receiving a score of 0 for Dice and a fixed penalty of 374 for HD95. The mean scores for the teams were calculated. Eight datasets comprising 1303 studies were annotated, with 402 studies (3076 lesions) released on Synapse as publicly available datasets to challenge competitors. Additionally, 31 studies (139 lesions) were held out for validation, and 59 studies (218 lesions) were used for testing. Segmentation accuracy was measured as rank across subjects, with the winning team achieving a LesionWise mean score of 7.9. The Dice score for the winning team was 0.65 ± 0.25. Common errors among the leading teams included false negatives for small lesions and misregistration of masks in space. The Dice scores and lesion detection rates of all algorithms diminished with decreasing tumor size, particularly for tumors smaller than 100 mm3. In conclusion, algorithms for BM segmentation require further refinement to balance high sensitivity in lesion detection with the minimization of false positives and negatives. The BraTS-METS 2023 challenge successfully curated well- annotated, diverse datasets and identified common errors, facilitating the translation of BM segmentation across varied environments and providing the tools for future development of personalized volumetric reports to patients undergoing BM treatment.

Background/Objectives: The Disabilities of the Arm, Shoulder, and Hand (DASH) questionnaire is a widely employed self-report tool for assessing upper extremity function. The aim of this study was to assess the psychometric properties of the Serbian version of the DASH by determining its criterion and construct validity, as well as internal consistency. Methods: This cross-sectional study was conducted among patients with hand and wrist disabilities at the Institute for Orthopedics “Banjica”, Serbia. The psychometric properties of the Serbian version of the DASH were analyzed through an examination of its factorial structure and internal consistency. The DASH consists of 30 items, 24 of which assess function, 21 of which focus on physical function and three on social/role function. The remaining six items evaluate symptoms related to pain, tingling/numbness, weakness, and stiffness. Results: A total of 297 patients were included in the study. The mean age was 47.4 ± 16.8 years, with 50.5% males. Three models were assessed to determine the reliability and validity of the questionnaire across different domains. Model 1 examined a single-factor structure. In Model 2, the items were divided into two domains: Physical Function and Psychosocial/Symptoms. In Model 3, items were subdivided into three domains: Physical Function, Symptoms, and Psychosocial. All models demonstrated an excellent internal consistency with a Cronbach’s alpha > 0.9 for most domains. The values for the fit indices Tucker–Lewis index (TLI) and Comparative-Fit Index (CFI) were above their cut-off criteria of 0.9, while the Root Mean Square Error of Approximation (RMSEA) and Standardized Root Mean Square Residual (SRMR) were below the suggested value of 0.06, indicating an excellent level of models fit. Standardized factor loadings were statistically significant (p < 0.05). Conclusions: The present study provided the evidence for the appropriate metric properties of the Serbian version of the DASH. Results support both the unidimensional and multidimensional structures of the DASH.