Evaluating robotic-assisted surgery training videos with multi-task convolutional neural networks

Author(s):  
Yihao Wang ◽  
Jessica Dai ◽  
Tara N. Morgan ◽  
Mohamed Elsaied ◽  
Alaina Garbens ◽  
...  
Author(s):  
Maria Castaldi ◽  
Mathias Palmer ◽  
Jorge Con ◽  
Ziad Abouezzi ◽  
Rifat Latifi ◽  
...  

Technology has had a dramatic impact on how diseases are diagnosed and treated. Although cut, sew, and tie remain the staples of surgical craft, new technical skills are required. While there is no replacement for live operative experience, training outside the operating room offers structured educational opportunities and stress modulation. A stepwise program for acquiring new technical skills required in robotic surgery involves three modules: ergonomic, psychomotor, and procedural. This is a prospective, educational research protocol aiming at evaluating the responsiveness of general surgery residents in Robotic-Assisted Surgery Training (RAST). Responsiveness is defined as change in performance over time. Performance is measured by the following content-valid metrics for each module. Module 1 proficiency in ergonomics includes: cart deploy, boom control, cart driving, camera port docking, targeting anatomy, flex joint, clearance joint and port nozzle adjusting, and routine and emergent undocking. Module 2 proficiency in psychomotor skills includes tissue handling, accuracy error, knot quality, and operating time. Module 3 proficiency in procedural skills prevents deviations from standardized sequential procedural steps in order to test length of specimen resection, angle for transection, vessel stump length post ligation, distance of anastomosis from critical landmarks, and proximal and distal resection margins. Resident responsiveness over time will be assessed comparing the results of baseline testing with final testing. Educational interventions will include viewing one instructional video prior to module commencement, response to module-specific multiple-choice questions, and individual weekly training sessions with a robotic instructor in the operating room. Residents will progress through modules upon successful final testing and will evaluate the educational environment with the Dundee Ready Educational Environment Measure (DREEM) inventory. The RAST program protocol outlined herein is an educational challenge with the primary endpoint to provide evidence that formal instruction has an impact on proficiency and safety in executing robotic skills.


Author(s):  
Wissam N. Raad ◽  
Adil Ayub ◽  
Chyun-Yin Huang ◽  
Landon Guntman ◽  
Sadiq S. Rehmani ◽  
...  

Objective Robotic-assisted surgery is increasingly being used in thoracic surgery. Currently, the Integrated Thoracic Surgery Residency Program lacks a standardized curriculum or requirement for training residents in robotic-assisted thoracic surgery. In most circumstances, because of the lack of formal residency training in robotic surgery, hospitals are requiring additional training, mentorship, and formal proctoring of cases before granting credentials to perform robotic-assisted surgery. Therefore, there is necessity for residents in Integrated Thoracic Surgery Residency Program to have early exposure and formal training on the robotic platform. We propose a curriculum that can be incorporated into such programs that would satisfy both training needs and hospital credential requirements. Methods We surveyed all 26 Integrated Thoracic Surgery Residency Program Directors in the United States. We also performed a PubMed literature search using the key word “robotic surgery training curriculum.” We reviewed various robotic surgery training curricula and evaluation tools used by urology, obstetrics gynecology, and general surgery training programs. We then designed a proposed curriculum geared toward thoracic Integrated Thoracic Surgery Residency Program adopted from our credentialing experience, literature review, and survey consensus. Results Of the 26 programs surveyed, we received 17 responses. Most Integrated Thoracic Surgery Residency Program directors believe that it is important to introduce robotic surgery training during residency. Our proposed curriculum is integrated during postgraduate years 2 to 6. In the preclinical stage postgraduate years 2 to 3, residents are required to complete introductory online modules, virtual reality simulator training, and in-house workshops. During clinical stage (postgraduate years 4–6), the resident will serve as a supervised bedside assistant and progress to a console surgeon. Each case will have defined steps that the resident must demonstrate competency. Evaluation will be based on standardized guidelines. Conclusions Expansion and utilization of robotic assistance in thoracic surgery have increased. Our proposed curriculum aims to enable Integrated Thoracic Surgery Residency Program residents to achieve competency in robotic-assisted thoracic surgery and to facilitate the acquirement of hospital privileges when they enter practice.


2020 ◽  
Vol 2020 (10) ◽  
pp. 28-1-28-7 ◽  
Author(s):  
Kazuki Endo ◽  
Masayuki Tanaka ◽  
Masatoshi Okutomi

Classification of degraded images is very important in practice because images are usually degraded by compression, noise, blurring, etc. Nevertheless, most of the research in image classification only focuses on clean images without any degradation. Some papers have already proposed deep convolutional neural networks composed of an image restoration network and a classification network to classify degraded images. This paper proposes an alternative approach in which we use a degraded image and an additional degradation parameter for classification. The proposed classification network has two inputs which are the degraded image and the degradation parameter. The estimation network of degradation parameters is also incorporated if degradation parameters of degraded images are unknown. The experimental results showed that the proposed method outperforms a straightforward approach where the classification network is trained with degraded images only.


Author(s):  
Edgar Medina ◽  
Roberto Campos ◽  
Jose Gabriel R. C. Gomes ◽  
Mariane R. Petraglia ◽  
Antonio Petraglia

Sign in / Sign up

Export Citation Format

Share Document