Mahajan, Amit Kumar

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.

OBJECTIVECushing disease is caused by a pituitary micro- or macroadenoma that hypersecretes adrenocorticotropic hormone (ACTH), resulting in hypercortisolemia. For decades, transsphenoidal resection (TSR) has been an efficacious treatment but with certain limitations, namely precise tumor localization and complete excision. The authors evaluated the novel use of a double-antibody sandwich assay for the real-time quantitation of ACTH in resected pituitary specimens with the goals of augmenting pathological diagnosis and ultimately improving long-term patient outcome.METHODSThis study involved a retrospective review of records and an analysis of assay values, pathology slides, and MRI studies of patients with Cushing disease who had undergone TSR in the period from 2009 to 2014 and had at least 1 year of follow-up in coordination with an endocrinologist. In the operating room, biopsy specimens from the patients had been analyzed for tissue ACTH concentration. Additional samples were simultaneously sent for frozen-section pathological analysis. The ACTH assay performance was compared against pathology assessments of surgical tumor samples using receiver operating characteristic (ROC) analysis and against pre- and postoperative MRI studies.RESULTSFourteen patients underwent TSR with guidance by ACTH-antibody assay and pathological assessment of 127 biopsy samples and were followed up for an average of 3 years. The ACTH threshold for discriminating adenomatous from normal tissue was 290,000 pg/mg of tissue, based on jointly maximized sensitivity (95.0%) and specificity (71.3%). Lateralization discordance between preoperative MRI studies and surgical visualization was noted in 3 patients, confirming the impression that MRI alone may not achieve optimal localization. A majority of the patients (85.7%) attained long-term disease remission based on urinary free cortisol levels, plasma cortisol levels, and long-term corticosteroid therapy. Comparisons of patient-months of remission and treatment failure showed that the remission rate in the study sample statistically exceeds the rate in historical controls (71.9%; p = 0.0007, Fisher’s exact test). Long-term unexpected hormonal deficiencies were statistically similar between study patients (29%) and those in a meta-analysis (25%; p = 0.7596, Fisher’s exact test).CONCLUSIONSThese preliminary findings reflect the promising potential of tissue-based ACTH-antibody–guided assay for improving the cure rates of Cushing disease patients undergoing TSR. Further studies with larger sample sizes, further refinements of assay interpretation, and longer-term follow-ups are needed.