Method for Patient-Specific Decision Support in Spine Surgery

Case ID:
C15678
Disclosure Date:
1/22/2019
Unmet Need
The current landscape of spine surgery suffers from unacceptably high variability in outcomes. Undesired events such as revision surgery, failed back surgery, and recurrent lumbar disc herniation are becoming increasingly common. The relationships between patient risk factors and outcomes are complex and poorly understood. Thus, determining the extent that a patient will benefit from surgery and recommending therapeutic options is a challenging and subjective task for spine surgeons operating within the current standard of care. Existing predictive models provide decision support using patient demographic information, not information from radiographic images. To the extent that image information is included, it is often manually determined and qualitative in nature. Tools that automatically analyze perioperative images and incorporate them into predictive models could drastically improve the performance of current models that only use patient demographic information.
 
Technology Overview
Johns Hopkins researchers have developed a framework to help support surgical decisions and predict spine surgery outcomes. This method extracts quantitative anatomical and morphological features of the patients’ spine from perioperative images and incorporates this information into current models that use patient demographics. This framework uses automatic image processing and machine learning algorithms to identify features that determine patient outcomes. These data can be used preoperatively to identify if the patient will benefit from surgery, and, if so, help select the optimal surgical approach. It can also be used post-operatively to provide prognostic information and guide rehabilitation. 
This new framework provides decision support to the surgeon and patient regarding various choices in treatment based on statistical analysis of patients who previously underwent surgery, whereas the current standard-of-care relies on subjective experience of the surgeon.
 
Stage of Development
Proof of concept.
 
Publications
Manuscript in preparation.
 
Patent Information:
Title App Type Country Serial No. Patent No. File Date Issued Date Expire Date Patent Status
IMAGE ANALYTICS FOR PREDICTIVE MODELING OF SURGICAL OUTCOMES PCT: Patent Cooperation Treaty European Patent Office 20765793.3   2/26/2020     Pending
DATA ANALYTICS FOR PREDICTIVE MODELING OF SURGICAL OUTCOMES PCT: Patent Cooperation Treaty United States 17/310,933   8/31/2021     Pending
Inventors:
Category(s):
Get custom alerts for techs in these categories/from these inventors:
For Information, Contact:
Lisa Schwier
lschwie2@jhu.edu
410-614-0300
Save This Technology:
2017 - 2022 © Johns Hopkins Technology Ventures. All Rights Reserved. Powered by Inteum