An AI model was developed for the automatic detection of pleomorphic lesions during capsule endoscopy, achieving notable results across various gastrointestinal sites. The binary esophagus CNN attained 83.6% accuracy, while the gastric CNN reached 96.6%. The small bowel CNN excelled with 97.6% accuracy in identifying lesions with different hemorrhagic potentials. Additionally, the colonic CNN demonstrated […]
Category: Digital Surgery and Telemedicine
Telementoring enhances surgical skills in rural hospitals
A surgical telementoring system significantly improved surgical skills and job interest among medical students in rural Japan. Surgeons reported enhanced mental performance and reduced frustration when using the telementoring system. Furthermore, a substantial increase in interest for rural hospital work was noted among medical students after the experience, with more expressing a desire to pursue […]
XR technologies enhance surgical training effectiveness and acceptance
An umbrella review of 44 studies highlights the significant potential of extended reality (XR) technologies in surgical training, particularly for orthopedics, neurology, and laparoscopy. Findings revealed that XR improves surgical skills and procedural accuracy while reducing risks and operating room time. User-friendly systems increased accessibility for trainees across skill levels, and positive reinforcement from experienced […]
Deep-learning model predicts hepatocellular carcinoma recurrence effectively
A novel deep-learning model, Recurr-Net, significantly outperformed traditional methods in predicting hepatocellular carcinoma (HCC) recurrence post-curative surgery. Analyzed across 1,231 patients, Recurr-Net demonstrated an area under the receiver operating characteristic curve (AUROC) of 0.770 to 0.857 internally and 0.758 to 0.798 externally, while the historical microvascular invasion (MVI) showed much lower performance. This indicates Recurr-Net’s […]
Deep learning model effectively predicts mismatch repair deficiency in colorectal cancer
A new deep learning model demonstrated strong predictive capabilities for mismatch repair deficiency in colorectal cancer. An area under the receiver operating characteristic (AUROC) score of 0.948 was achieved during testing, while independent validation yielded an AUROC of 0.807. Notably, the model achieved a negative predictive value of 94.2%. In a prospective trial, it accurately […]
AI system significantly improves surgical case length predictions
An artificial intelligence system was developed for predicting surgical case lengths, significantly outperforming current electronic health record estimates. Analyzed data from 125,493 inpatient elective surgical cases revealed the categoricalboost regressor using clinical text embeddings achieved a mean absolute error of 46.4 minutes, compared to 120.0 minutes for existing records (p < 0.001). The AI model […]
AI surpasses human endoscopists in detecting colorectal lesions
A deep learning algorithm based on the yolov7 model demonstrated exceptional performance in identifying laterally spreading tumors (LSTs) with an accuracy of 99.34%. Sensitivity was recorded at 96.88%, while specificity reached 99.8%. This algorithm outperformed both novice and senior endoscopists, achieving diagnostic capabilities comparable to expert professionals. The study utilized a substantial dataset of 7985 […]
Decentralized Knowledge Graph Improves Medical Data Organization
A decentralized knowledge graph (dkg) was developed to enhance the organization of medical and surgical data, improving the quality of health data collection. By utilizing the cyb.ai protocol within the cosmos network, this system leverages emergency surgery publications to create a structured bibliographic resource. The dkg not only facilitates a better understanding of medical relationships […]
New segmentation method significantly improves liver cancer imaging
The DSA-Former network introduces a hybrid approach for segmenting liver and liver tumours, effectively integrating convolutional kernels and attention mechanisms. Achieving dice coefficients of 96.8% for liver segmentation and 72.2% for tumour segmentation, this method outperforms existing techniques in key metrics such as IoU and HD95. Enhanced segmentation precision promises to bolster the accuracy of […]
Postoperative complications in colorectal cancer surgery lack uniform guidelines
A systematic review of 135 articles on postoperative complications from colorectal cancer surgery underscores significant disparities in research focus and treatment guidelines. Notably, anastomotic leakage and postoperative infections accounted for 35% and 25% of studies, while port-site metastases and cardiovascular dysfunctions received limited attention. The prevalence rates of complications varied widely, from 3% to 20%, […]