Accurate preoperative prediction of lymphovascular and perineural invasion status in gastric cancer patients is crucial for identifying high-risk individuals. This study developed and tested a machine learning model that successfully combines spectral CT parameters and clinical indicators to accurately predict LVI/PNI status. The integration of portal venous and ep spectral CT parameters significantly enhances the […]
Category: Digital Surgery and Telemedicine
GPT-4 Demonstrates Superior Readability in Bariatric Surgery Patient Education
Large language models (LLMs) like GPT-4 significantly improve the readability of bariatric surgery patient education materials (PEMs), with responses ranging from 6th to 9th grade reading levels. While GPT-3.5 and GPT-4 showed similar accuracy and comprehensiveness in simplifying responses, Bard’s responses were less comprehensive when simplified. This study emphasizes the potential of LLMs in enhancing […]
Virtual Reality with Head-Mounted Display Improves Decision-Making in Hepatic Surgery Training.
A study compared virtual reality with head-mounted display (VR-HMD), on-screen 3D visualization, and 2D computed tomography for hepatic surgery training. Results showed higher accuracy and faster responses in the VR-HMD and 3D groups compared to the 2D group. Both VR-HMD and 3D were equally effective, indicating VR-HMD’s usefulness in determining liver anatomy and tumor involvement […]
Preoperative History and Physical Update Visits: Minimal Impact on Operative Management
Researchers evaluated the impact of preoperative history and physical update visits on 8,683 patients across different surgical specialties in 2019. They found that the majority of documented changes were in histories, with minimal impact on physical exams and operative plans. 99.2% of visits were suitable for telehealth, suggesting a potential alternative to in-person visits. The […]
Advancements in Surgical Visualization and Guidance Technologies
Recent advancements in surgical visualization and guidance, including fluorescence-guided surgery, contrast-enhanced ultrasound, and augmented reality, are shaping the future of general surgery. This narrative review explores cutting-edge technologies and their integration with artificial intelligence, promising enhanced intraoperative decision-making and surgical precision. While these innovations offer exciting prospects, further research and surgeon training in artificial intelligence […]
Assessment of AI-generated Medical Information on Appendicitis
Generative AI chatbots like ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2 were assessed for content and quality on appendicitis information. Results showed favorable quality scores for all but Claude-2, which had significantly lower quality. Bard was the only platform providing verifiable sources, while others advised consulting a physician. Readability levels exceeded public standards, indicating potential for patient […]
Virtual Informed Consent and the Importance of Trust-Building
Virtual formats for informed consent in surgery may hinder trust-building, as they rely on an information-transfer model rather than a trust-building one. This study suggests that a trust-building model is essential for a fuller understanding of the consent process, highlighting potential shortcomings of virtual formats on interpersonal and systemic levels. The ethical framework proposed can […]
Development and Validation of a Privacy-proof Surgical Anonymization Algorithm for Live Streaming
Researchers developed and validated a pioneer surgical anonymization algorithm, named Robotic Anonymization Network (ROBAN), for real-time removal of out-of-body images during live surgery streaming. The algorithm achieved a high ROC AUC score of 99.89% after post-processing, outperforming previous state-of-the-art methods and offering reliable, accurate, and safe anonymization across different robotic platforms and procedural types. Journal […]
Development of Prediction Models for Mortality Risk After Colorectal Cancer Surgery
Researchers developed prediction models using preoperative data to assess short-term mortality risk following colorectal cancer surgery. The models showed good discrimination and calibration, with an area under the receiver operating characteristic curve of 0.88, 0.878, and 0.861 for 30-day, 90-day, and 1-year mortality, respectively. Subgroup analysis did not improve discrimination or calibration. Combining all operated […]
Machine learning does not outperform expert-informed predictions for outcome after major liver surgery
Expert-informed prediction models and machine learning models both showed similar performance in predicting postoperative overall morbidity in liver surgery patients, with high variability and over/underprediction. There was no improvement in performance for major liver surgery patients, indicating that machine learning may not have clinical relevance for predicting postoperative outcomes at this stage. Journal Article by […]