Significance Omission of steps, wrong sequencing and excessive force in laparoscopic surgery lead to unnecessary suffering and costly litigation. Due to rapidly advancing techniques, technology and work hour restrictions, only a fraction of the full spectrum of lap procedures and safety concerns can be addressed by current basic dexterity training or by 'see-one-do-one' in the OR. Computer-based training with force-feedback promises to help fill the gap. But commercial products focus on standard cases and do not allow surgeon-educators themselves to select and fine-tune anatomy, pathology and technique. Innovation The proposed authoring environment will broaden the range, deepen the specificity and speed up innovation in computer-based training by enabling surgeon-educators (a) to define their focus and scope via a structured list of surgical steps from which module components are automatically initialized; (b) to fine-tune crucial details that convey surgical expertise, techniqe and insight. This enables growing a rich collection of sharable training modules across a variety of surgical procedures. No similar approach currently exists. - Building on state-of-the-art solutions of tissue physics, GPU-computing and graphics challenges of interactive 3D surgery simulation with force-feedback, this proposal additionally breaks new ground for simulation technology by (a) automatically instantiating modules (instruction pages, VR-scenario, measurement) and (b) enabling surgeons to fine-tune scenarios. Approach (1) For each anatomic/pathological variant of appendectomy, cholecystectomy and gastric bypass, a lead surgeon will draft, and at least three surgeons will critique, a web-based step-by-step list of task and safety issues. In the rare case when no consensus is reached, competing lists and cases are generated. Heads of residency, chief surgeons and a dean for Simulation and Medical Education have signed up to lead the efforts at five institutions. (2) The output, a highly structured list of task and safety issues, enables software to automatically initialize a training module's instructional pages and VR-scenario. This raw module is fine-tuned by the lead author who also sets acceptable performance ranges by executing the VR-simulation. The resulting modules undergo a peer-review cycle. VR-simulation leverages a simulation engine and an extensive database of anatomy and surgical tools developed under R21 funding, published and demonstrated at conferences [35, 37, 34, 14, 57, 58, 79, 56, 44, 36, 76]. (3) The impact of each released module on safety awareness in the OR is measured by a separate small randomized, blinded study at two of the five participating medical centers. (4) The authoring environment and its results will be advertised at major medical conferences (AAMC, ACS CC, ACGME) and its structured code base will be disseminated on the web under the GNU LGP License to encourage broad sharing and enable modification and continued distributed development supported by the stakeholders.

Public Health Relevance

To improve competency and safety-awareness of laparoscopy trainees, we provide and validate a software environment for authoring simulation-based Virtual Reality training modules. This environment empowers surgeon-educators to define the modules' focus and scope, automatically initialize all components of the module and to, by themselves, fine-tune crucial details that convey surgical expertise. This approach encourages growing a rich collection of VR training modules across a variety of surgical procedures.

Agency
National Institute of Health (NIH)
Institute
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Type
Research Project (R01)
Project #
5R01EB018625-03
Application #
9032997
Study Section
Biomedical Library and Informatics Review Committee (BLR)
Program Officer
Peng, Grace
Project Start
2014-03-10
Project End
2018-02-28
Budget Start
2016-03-01
Budget End
2017-02-28
Support Year
3
Fiscal Year
2016
Total Cost
Indirect Cost
Name
University of Florida
Department
Engineering (All Types)
Type
Biomed Engr/Col Engr/Engr Sta
DUNS #
969663814
City
Gainesville
State
FL
Country
United States
Zip Code
32611
Kurenov, Sergei; Cendan, Juan; Dindar, Saleh et al. (2017) Surgeon-Authored Virtual Laparoscopic Adrenalectomy Module Is Judged Effective and Preferred Over Traditional Teaching Tools. Surg Innov 24:72-81
Nguyen, Dang-Manh; Peters, Jörg (2017) EXPLICIT LEAST-DEGREE BOUNDARY FILTERS FOR DISCONTINUOUS GALERKIN. SIAM J Sci Comput 39:A1741-A1765
Kar?iauskas, K?stutis; Peters, Jörg (2017) Improved shape for refinable surfaces with singularly parameterized irregularities. Comput Aided Des 90:191-198
Nguyen, Thien; Peters, Jörg (2016) Refinable C(1) spline elements for irregular quad layout. Comput Aided Geom Des 43:123-130
Dindar, Saleh; Nguyen, Thien; Peters, Jörg (2016) Towards surgeon-authored VR training: the scene-development cycle. Stud Health Technol Inform 220:103-9
Nguyen, Thien; Kar?iauskas, K?stutis; Peters, Jörg (2016) C(1) finite elements on non-tensor-product 2d and 3d manifolds. Appl Math Comput 272:148-158
Groisser, David; Peters, Jörg (2015) Matched G(k) -constructions always yield C(k) -continuous isogeometric elements. Comput Aided Geom Des 34:67-72
Peters, Jörg (2015) General spline filters for discontinuous Galerkin solutions. Comput Math Appl 70:1046-1050