Predicting decompression surgery by applying multimodal deep learning to patients’ structured and unstructured health data

BMC Medical Informatics and Decision Making

Research

Eric Meier¹  Laura S. Gold²  Jeffrey G. Jarvik³  Janna Friedly⁴  Pradeep Suri⁴  Sean D. Mooney⁵  Trevor Cohen⁵  Chethan Jujjavarapu⁵  Patrick J. Heagerty⁶  Vikas Pejaver⁷ 

[1] Clinical Learning, Evidence and Research Center, University of Washington, 4333 Brooklyn Ave NE, 98105, Seattle, WA, USA;Department of Biostatistics, University of Washington, Box 357232, 98195-7232, Seattle, WA, USA;Center for Biomedical Statistics, University of Washington, Seattle, WA, USA;Clinical Learning, Evidence and Research Center, University of Washington, 4333 Brooklyn Ave NE, 98105, Seattle, WA, USA;Department of Radiology, University of Washington, 1959 NE Pacific Street, 98195, Seattle, WA, USA;Clinical Learning, Evidence and Research Center, University of Washington, 4333 Brooklyn Ave NE, 98105, Seattle, WA, USA;Department of Radiology, University of Washington, 1959 NE Pacific Street, 98195, Seattle, WA, USA;Department of Neurological Surgery, University of Washington, 1959 NE Pacific Street, 98195, Seattle, WA, USA;Department of Health Services, University of Washington, Box 357660, 98195-7660, Seattle, WA, USA;Clinical Learning, Evidence and Research Center, University of Washington, 4333 Brooklyn Ave NE, 98105, Seattle, WA, USA;Department of Rehabilitation Medicine, University of Washington, 1959 NE Pacific St, 98195, Seattle, WA, USA;Department of Biomedical Informatics and Medical Education, School of Medicine, University of Washington, Box 358047, 98195, Seattle, WA, USA;Department of Biostatistics, University of Washington, Box 357232, 98195-7232, Seattle, WA, USA;Center for Biomedical Statistics, University of Washington, Seattle, WA, USA;Institute for Genomic Health, Icahn School of Medicine at Mount Sinai, 10029, New York, NY, USA;Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, 10029, New York, NY, USA;

关键词: Lower back pain;    Lumbar spinal stenosis;    Lumbar disc herniation;    Deep learning;    Generalizability;    Multimodal;    Machine learning;    Decompression surgery;    Prediction;    Classification;   

DOI  :  10.1186/s12911-022-02096-x

 received in 2022-05-30, accepted in 2022-12-29,  发布年份 2022

来源: Springer

【 摘 要 】

BackgroundLow back pain (LBP) is a common condition made up of a variety of anatomic and clinical subtypes. Lumbar disc herniation (LDH) and lumbar spinal stenosis (LSS) are two subtypes highly associated with LBP. Patients with LDH/LSS are often started with non-surgical treatments and if those are not effective then go on to have decompression surgery. However, recommendation of surgery is complicated as the outcome may depend on the patient’s health characteristics. We developed a deep learning (DL) model to predict decompression surgery for patients with LDH/LSS.Materials and methodWe used datasets of 8387 and 8620 patients from a prospective study that collected data from four healthcare systems to predict early (within 2 months) and late surgery (within 12 months after a 2 month gap), respectively. We developed a DL model to use patients’ demographics, diagnosis and procedure codes, drug names, and diagnostic imaging reports to predict surgery. For each prediction task, we evaluated the model’s performance using classical and generalizability evaluation. For classical evaluation, we split the data into training (80%) and testing (20%). For generalizability evaluation, we split the data based on the healthcare system. We used the area under the curve (AUC) to assess performance for each evaluation. We compared results to a benchmark model (i.e. LASSO logistic regression).ResultsFor classical performance, the DL model outperformed the benchmark model for early surgery with an AUC of 0.725 compared to 0.597. For late surgery, the DL model outperformed the benchmark model with an AUC of 0.655 compared to 0.635. For generalizability performance, the DL model outperformed the benchmark model for early surgery. For late surgery, the benchmark model outperformed the DL model.ConclusionsFor early surgery, the DL model was preferred for classical and generalizability evaluation. However, for late surgery, the benchmark and DL model had comparable performance. Depending on the prediction task, the balance of performance may shift between DL and a conventional ML method. As a result, thorough assessment is needed to quantify the value of DL, a relatively computationally expensive, time-consuming and less interpretable method.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

Fig. 2

Fig. 54

Fig. 2

Fig. 1

12888_2022_4500_Article_IEq2.gif

Fig. 2