Graduate Seminar Series: Cell and Tissue Stream – Johnathan Lex

Graduate Seminar Series: Cell and Tissue Stream
Graduate Seminar Series for the Institute of Biomedical Engineering (BME). This day is for cell and tissue stream presenters.
If you would like to invite your Principal Investigator, please add their email via the ‘Add Guest’ button and they will also be notified of your presentation.
Presentation Title: Machine Learning to Predict and Optimize Arthroplasty Resource Utilization and Patient Outcomes
Abstract:
Background
The gold-standard therapy for end-stage arthritis of the knee and hip is total knee and hip arthroplasty (TKA and THA, respectively). They are the most common orthopaedic procedures carried out in the US, with over 1.3 million of these operations performed in 2018, a 21.9% increase from 2008. Global ageing and the continued obesity epidemic will cause this figure to continue to rise. Inefficiencies in surgical services are a major cause of Canadian healthcare’s growing costs and wait times. OR scheduling is a critical consideration in surgical care delivery with direct impacts on staffing and inventory planning across multiple hospital units (post- operative care units, inpatient wards, intensive care units). Manual scheduling requires that the human planner simultaneously reason about unknowns such as case-specific length-of-surgery, hard constraints on the total amount of available OR time, surgeon availability and caseload, and soft constraints such as patient wait times, case priorities, and post-operative quality of life improvement, all while maximizing throughput.
Specific Aims
1. To evaluate neural network prediction models incorporating patient and operational factors for the prediction of duration of surgery and length of stay for primary and revision TKA/THA.
2. To compare the impact of a two-stage machine learning and optimization approach on operating room efficiency compared to using historic scheduling practices (surgeon- estimated DOS), mean DOS, mean surgeon-specific DOS and rolling mean surgeon- specific DOS.
3. Determine if neural networks with patient, surgeon and implant features can accurately predict improvement in patient reported outcome measures following TKA/THA.
Supervisor Name: Dr. Cari Whyne
Year of Study: 2
Program of Study: MASc
Zoom link: https://us02web.zoom.us/j/89610372821?pwd=azd4SCtYVWtreVovaGNPV1c2NGY2Zz09
Meeting ID: 896 1037 2821
Password: 483329
Powered by Calendly.com