Tag: microvascular anastomosis

Posted on

Artificial intelligence–based deep learning model for evaluating procedural consistency in microvascular anastomosis

J Neurosurg 144:1–10, 2026

This study presents an LSTM-based deep learning model that objectively evaluates microvascular anastomosis performance by predicting hand-motion trajectories from MediaPipe-derived hand landmarks. It quantifies consistency using Kullback-Leibler divergence and validates complementary metrics—economy and flow of motion—comparing two expert neurosurgeons (repeat sessions) and one trainee in simulated end-to-side anastomoses.

Results show low KL divergence for experts versus higher divergence for the trainee, reflecting greater consistency and efficiency. The authors discuss methodology, model architecture choices, limitations in generalizability, and potential integration into microsurgical training workflows for objective skill assessment.

Deep Learning Model: An LSTM-based neural network was developed to objectively assess consistency and precision in microvascular anastomosis by predicting and comparing suturing hand movements using video-based hand landmark tracking, eliminating the need for physical sensors.

Hand Tracking Technology: The model utilized MediaPipe Hand Landmarker, a CNN-based system that detects 21 hand landmarks from standard video, enabling detailed, sensor-free motion analysis during microsurgical simulation.

Performance Metrics: Three primary metrics were used: Kullback-Leibler (KL) divergence for consistency, economy of motion (mean Euclidean distance of hand movement), and flow of motion (median time per suture), providing quantitative, objective evaluation of surgical skill.

Experimental Setup: Two expert neurosurgeons performed microanastomosis simulations (interrupted and continuous suturing) in two sessions one year apart, and a trainee performed the same task for comparison; all sessions were recorded and analyzed using the AI pipeline.

Results and Interpretation: Experts showed low KL divergence (high consistency) and efficient, rhythmic motion, while the trainee had higher KL divergence, longer suture intervals, and more variable motion, reflecting less developed skill.

Model Application: The approach enables rapid, automated assessment of multiple trainees using standard video equipment, supporting objective tracking of skill progression and facilitating feedback in training environments.

Model Rationale: LSTM architecture was chosen for its ability to model long-term temporal dependencies in sequential hand movement data, making it suitable for predicting surgical motion patterns over extended timeframes.

Limitations and Future Directions: Current findings are based on a small sample of experts and one trainee in a simulated environment; broader validation, metric standardization (especially for KL divergence), and extension to real operative settings are needed for generalizability.

Posted on

A Novel Approach for Free, Affordable, and Sustainable Microsurgery Laboratory Training for Low- and Middle-Income Countries: University of Wisconsin-Madison Microneurosurgery Laboratory Experience

Neurosurgery 94:1311–1323, 2024

In low- and middle-income countries (LMICs), approximately 5 million essential neurosurgical operations per year remain unaddressed. When compared with high-income countries, one of the reasons for this disparity is the lack of microsurgery training laboratories and neurosurgeons trained in microsurgical techniques. In 2020, we founded the Madison Microneurosurgery Initiative to provide no-cost, accessible, and sustainable microsurgery training opportunities to health care professionals from LMICs in their respective countries.

METHODS: We initially focused on enhancing our expertise in microsurgery laboratory training requirements. Subsequently, we procured a wide range of stereo microscopes, light sources, and surgical instrument sets, aiming to develop affordable, high-quality, and long-lasting microsurgery training kits. We then donated those kits to neurosurgeons across LMICs. After successfully delivering the kits to designated locations in LMICs, we have planned to initiate microsurgery laboratory training in these centers by providing a combination of live-streamed, offline, and in-person training assistance in their institutions.

RESULTS: We established basic microsurgery laboratory training centers in 28 institutions across 18 LMICs. This was made possible through donations of 57 microsurgery training kits, including 57 stereo microscopes, 2 surgical microscopes, and several advanced surgical instrument sets. Thereafter, we organized 10 live-streamed microanastomosis training sessions in 4 countries: Lebanon, Paraguay, Türkiye, and Bangladesh. Along with distributing the recordings from our live-streamed training sessions with these centers, we also granted them access to our microsurgery training resource library. We thus equipped these institutions with the necessary resources to enable continued learning and hands-on training. Moreover, we organized 7 in-person no-cost hands-on microanastomosis courses in different institutions across Türkiye, Georgia, Azerbaijan, and Paraguay. A total of 113 surgical specialists successfully completed these courses.

CONCLUSION: Our novel approach of providing microsurgery training kits in combination with live-streamed, offline, and in-person training assistance enables sustainable microsurgery laboratory training in LMICs.

Posted on

Seven bypasses simulation set: description and validity assessment of novel models for microneurosurgical training

J Neurosurg 138:732–739, 2023

Microsurgical training remains indispensable to master cerebrovascular bypass procedures, but simulation models for training that accurately replicate microanastomosis in narrow, deep-operating corridors are lacking. Seven simulation bypass scenarios were developed that included head models in various surgical positions with premade approaches, simulating the restrictions of the surgical corridors and hand positions for microvascular bypass training. This study describes these models and assesses their validity.

METHODS Simulation models were created using 3D printing of the skull with a designed craniotomy. Brain and external soft tissues were cast using a silicone molding technique from the clay-sculptured prototypes. The 7 simulation scenarios included: 1) temporal craniotomy for a superficial temporal artery (STA)–middle cerebral artery (MCA) bypass using the M4 branch of the MCA; 2) pterional craniotomy and transsylvian approach for STA-M2 bypass; 3) bifrontal craniotomy and interhemispheric approach for side-to-side bypass using the A3 branches of the anterior cerebral artery; 4) far lateral craniotomy and transcerebellomedullary approach for a posterior inferior cerebellar artery (PICA)–PICA bypass or 5) PICA reanastomosis; 6) orbitozygomatic craniotomy and transsylvian-subtemporal approach for a posterior cerebral artery bypass; and 7) extended retrosigmoid craniotomy and transcerebellopontine approach for an occipital artery–anterior inferior cerebellar artery bypass. Experienced neurosurgeons evaluated each model by practicing the aforementioned bypasses on the models. Face and content validities were assessed using the bypass participant survey.

RESULTS A workflow for model production was developed, and these models were used during microsurgical courses at 2 neurosurgical institutions. Each model is accompanied by a corresponding prototypical case and surgical video, creating a simulation scenario. Seven experienced cerebrovascular neurosurgeons practiced microvascular anastomoses on each of the models and completed surveys. They reported that actual anastomosis within a specific approach was well replicated by the models, and difficulty was comparable to that for real surgery, which confirms the face validity of the models. All experts stated that practice using these models may improve bypass technique, instrument handling, and surgical technique when applied to patients, confirming the content validity of the models.

CONCLUSIONS The 7 bypasses simulation set includes novel models that effectively simulate surgical scenarios of a bypass within distinct deep anatomical corridors, as well as hand and operator positions. These models use artificial materials, are reusable, and can be implemented for personal training and during microsurgical courses.

Proudly powered by WordPress