Deep learning system for malignancy risk prediction in cystic renal lesions: a multicenter study.

OBJECTIVES: To develop an interactive, non-invasive artificial intelligence (AI) system for malignancy risk prediction in cystic renal lesions (CRLs). METHODS: In this retrospective, multicenter diagnostic study, we evaluated 715 patients. An interactive geodesic-based 3D segmentation model was created for CRLs segmentation. A CRLs classification model was developed using spatial encoder temporal decoder (SETD) architecture. The classification model combines a 3D-ResNet50 network for extracting spatial features and a gated recurrent unit (GRU) network for decoding temporal features from multi-phase CT images. We assessed the segmentation model using sensitivity (SEN), specificity (SPE), intersection over union (IOU), and dice similarity (Dice) metrics. The classification model's performance was evaluated using the area under the receiver operator characteristic curve (AUC), accuracy score (ACC), and decision curve analysis (DCA). RESULTS: From 2012 to 2023, we included 477 CRLs (median age, 57 [IQR: 48-65]; 173 men) in the training cohort, 226 CRLs (median age, 60 [IQR: 52-69]; 77 men) in the validation cohort, and 239 CRLs (median age, 59 [IQR: 53-69]; 95 men) in the testing cohort (external validation cohort 1, cohort 2, and cohort 3). The segmentation model and SETD classifier exhibited excellent performance in both validation (AUC = 0.973, ACC = 0.916, Dice = 0.847, IOU = 0.743, SEN = 0.840, SPE = 1.000) and testing datasets (AUC = 0.998, ACC = 0.988, Dice = 0.861, IOU = 0.762, SEN = 0.876, SPE = 1.000). CONCLUSION: The AI system demonstrated excellent benign-malignant discriminatory ability across both validation and testing datasets and illustrated improved clinical decision-making utility. CRITICAL RELEVANCE STATEMENT: In this era when incidental CRLs are prevalent, this interactive, non-invasive AI system will facilitate accurate diagnosis of CRLs, reducing excessive follow-up and overtreatment. KEY POINTS: The rising prevalence of CRLs necessitates better malignancy prediction strategies. The AI system demonstrated excellent diagnostic performance in identifying malignant CRL. The AI system illustrated improved clinical decision-making utility.

Reconstruction of a massive chest wall defect using a free anterolateral-lower medial thigh flaps: a case report.

Critical defects of the chest wall require robust soft tissue coverage to protect the thoracic viscera. We define massive chest wall defects as larger than two-thirds of the chest wall. For such defects, classic flaps like the omentum, latissimus dorsi and anterolateral thigh flaps are usually insufficient. In our patient, a bilateral total mastectomy for locally advanced breast cancer resulted in a massive chest wall defect (40 by 30 cm). Soft tissue coverage was achieved with a combined anterolateral-lower medial thigh flaps. Revascularization of the anterolateral thigh and lower medial thigh components was via the internal mammary and thoracoacromial vessels, respectively. Post-operative recovery was uneventful and the patient received adjuvant chemoradiotherapy in a timely manner. The total follow up was 24-months. We illustrate the novel use of the lower medial thigh territory in extending the size of the anterolateral thigh flap to reconstruct massive chest wall defects.

A Preliminary Experience of Endoscopic Total Mastectomy With Immediate Free Abdominal-Based Perforator Flap Reconstruction Using Minimal Incisions, and Literature Review.

PURPOSE: Endoscopic total mastectomy (ETM) is predominantly performed with reconstruction using prostheses, lipofilling, omental flaps, latissimus dorsi flaps, or a combination of these techniques. Common approaches include minimal incisions, e.g., periareolar, inframammary, axillary, or mid-axillary line, which limit the technical ability to perform autologous flap insets and microvascular anastomoses, as such the ETM with free abdominal-based perforator flap reconstruction has not been robustly explored. METHODS: We studied female patients with breast cancer who underwent ETM and abdominal-based flap reconstruction. Clinical-radiological-pathological characteristics, surgery, complications, recurrence rates, and aesthetic outcomes were reviewed. RESULTS: Twelve patients underwent ETM with abdominal-based flap reconstruction. The mean age was 53.4 years (range 36-65). Of the patients, 33.3% were surgically treated for stage I, 58.4% for stage II, and 8.3% for stage III cancer. Mean tumor size was 35.4 mm (range 1-67). Mean specimen weight was 458.75 g (range 242-800). Of the patients, 92.3% successfully received endoscopic nipple-sparing mastectomy and 7.7% underwent intraoperative conversion to skin-sparing mastectomy after carcinoma was reported on frozen section of the nipple base. Mean operative time for ETM was 139 minutes (92-198), and the average ischemic time was 37.3 minutes (range 22-50). Fifty percent of patients underwent deep inferior epigastric perforator, 33.4% underwent MS-2 transverse rectus abdominis musculocutaneous (TRAM), 8.3% underwent MS-1 TRAM, and 8.3% underwent pedicled TRAM flap reconstruction. No cases required re-exploration, no flap failure occurred, margins were clear, and no skin or nipple-areolar complex ischemia/necrosis developed. In the aesthetic outcome evaluation, 16.7% were excellent, 75% good, 8.3% fair, and none were unsatisfactory. No recurrences were observed. CONCLUSION: ETM through a minimal-access inferior mammary or mid-axillary line approach, followed by immediate pedicled TRAM or free abdominal-based perforator flap reconstruction, can be a safe means of achieving an "aesthetically scarless" mastectomy and reconstruction through minimal incisions.

Deep learning and radiomic feature-based blending ensemble classifier for malignancy risk prediction in cystic renal lesions.

BACKGROUND: The rising prevalence of cystic renal lesions (CRLs) detected by computed tomography necessitates better identification of the malignant cystic renal neoplasms since a significant majority of CRLs are benign renal cysts. Using arterial phase CT scans combined with pathology diagnosis results, a fusion feature-based blending ensemble machine learning model was created to identify malignant renal neoplasms from cystic renal lesions (CRLs). Histopathology results were adopted as diagnosis standard. Pretrained 3D-ResNet50 network was selected for non-handcrafted features extraction and pyradiomics toolbox was selected for handcrafted features extraction. Tenfold cross validated least absolute shrinkage and selection operator regression methods were selected to identify the most discriminative candidate features in the development cohort. Feature's reproducibility was evaluated by intra-class correlation coefficients and inter-class correlation coefficients. Pearson correlation coefficients for normal distribution and Spearman's rank correlation coefficients for non-normal distribution were utilized to remove redundant features. After that, a blending ensemble machine learning model were developed in training cohort. Area under the receiver operator characteristic curve (AUC), accuracy score (ACC), and decision curve analysis (DCA) were employed to evaluate the performance of the final model in testing cohort. RESULTS: The fusion feature-based machine learning algorithm demonstrated excellent diagnostic performance in external validation dataset (AUC = 0.934, ACC = 0.905). Net benefits presented by DCA are higher than Bosniak-2019 version classification for stratifying patients with CRL to the appropriate surgery procedure. CONCLUSIONS: Fusion feature-based classifier accurately distinguished malignant and benign CRLs which outperformed the Bosniak-2019 version classification and illustrated improved clinical decision-making utility.

Biobrane™ for burns of the pubic region: minimizing dressing changes.

BACKGROUND: The pubic region is often involved in accidental hot water or soup-spill burns. Most of these wounds are superficial partial thickness burns. Due to their proximity to the urinary system, as well as vaginal and anal openings, these burns are easily contaminated. Daily dressings are routinely prescribed as the sole treatment. The cumbersome dressing process is uncomfortable and embarrassing for patients. Biobrane™ is a bilayered biosynthetic dressing. Its coverage of superficial partial thickness burns promotes wound healing and allows one-time application. CASE PRESENTATIONS: We report two patients who suffered superficial dermal burns over their pubic region. One patient had 23% total body surface area (TBSA) burns over her lower abdomen, both thighs and pubic region. The second patient had 10% TBSA burns that involved her perineum and the medial sides of both thighs and buttocks. Both were managed with the standard resuscitation protocol in the initial phase. Their burn injuries were managed by shaving, Foley catheterization and Biobrane™ coverage. Their wounds healed uneventfully without complications. Full epithelization was achieved by post-operative day seven. Both patients consented to medical photography and academic publication. CONCLUSION: Shaving and catheterization improved the hygiene of the burns of the pubic area. The Biobrane™ method circumvents the need of regular dressing changes, eliminating the pain due to dressing changes and preserving patient dignity.