Stanford Neurosurgical
Artificial Intelligence and Machine Learning Laboratory

The safety and efficacy of surgery is our number one priority. As we enter the era of big data, the onus is upon us to utilize what we have learned to improve medical care for future generations. Using novel, cutting edge artificial intelligence and machine learning techniques, our goal is to mine through millions of patient records to predict outcomes following all types of surgery. Imagine a virtual algorithm, that is capable of providing a true estimate of surgical risk, outcome, and efficacy – based specifically on your characteristics. This is the very foundation of precision medicine, and will empower physicians to deliver outstanding care.


The focus of my laboratory is to utilize precision medicine techniques to improve the diagnosis and treatment of neurologic conditions. From traumatic brain injury to spinal scoliosis, the ability to capture detailed data regarding clinical symptoms and treatment outcomes has empowered us to do better for patients. Utilize data to do better for patients, that’s what we do.

Assistant Professor of Neurosurgery and, by courtesy, of Orthopaedic Surgery at the Stanford University Medical Center


  • Patterns of Opioid and Benzodiazepine Use in Opioid-Naive Patients with Newly Diagnosed Low Back and Lower Extremity Pain. Journal of general internal medicine Azad, T. D., Zhang, Y., Stienen, M. N., Vail, D., Bentley, J. P., Ho, A. L., Fatemi, P., Herrick, D., Kim, L. H., Feng, A., Varshneya, K., Jin, M., Veeravagu, A., Bhattacharya, J., Desai, M., Lembke, A., Ratliff, J. K. 2019


    BACKGROUND: The morbidity and mortality associated with opioid and benzodiazepine co-prescription is a pressing national concern. Little is known about patterns of opioid and benzodiazepine use in patients with acute low back pain or lower extremity pain.OBJECTIVE: To characterize patterns of opioid and benzodiazepine prescribing among opioid-naive, newly diagnosed low back pain (LBP) or lower extremity pain (LEP) patients and to investigate the relationship between benzodiazepine prescribing and long-term opioid use.DESIGN/SETTING: We performed a retrospective analysis of a commercial database containing claims for more than 75 million enrollees in the USA.PARTICIPANTS: Participants were adult patients newly diagnosed with LBP or LEP between 2008 and 2015 who did not have a red flag diagnosis, had not received an opioid prescription in the 6months prior to diagnosis, and had 12months of continuous enrollment after diagnosis.MAIN OUTCOMES AND MEASURES: Among patients receiving at least one opioid prescription within 12months of diagnosis, we defined discrete patterns of benzodiazepine prescribing-continued use, new use, stopped use, and never use. We tested the association of these prescription patterns with long-term opioid use, defined as six or more fills within 12months.RESULTS: We identified 2,497,653 opioid-naive patients with newly diagnosed LBP or LEP. Between 2008 and 2015, 31.9% and 11.5% of these patients received opioid and benzodiazepine prescriptions, respectively, within 12months of diagnosis. Rates of opioid prescription decreased from 34.8% in 2008 to 27.0% in 2015 (P<0.001); however, prescribing of benzodiazepines only decreased from 11.6% in 2008 to 10.8% in 2015. Patients with continued or new benzodiazepine use consistently used more opioids than patients who never used or stopped using benzodiazepines during the study period (one-way ANOVA, P<0.001). For patients with continued and new benzodiazepine use, the odds ratio of long-term opioid use compared with those never prescribed a benzodiazepine was 2.99 (95% CI, 2.89-3.08) and 2.68 (95% CI, 2.62-2.75), respectively.LIMITATIONS: This study used administrative claims analyses, which rely on accuracy and completeness of diagnostic, procedural, and prescription codes.CONCLUSION: Overall opioid prescribing for low back pain or lower extremity pain decreased substantially during the study period, indicating a shift in management within the medical community. Rates of benzodiazepine prescribing, however, remained at approximately 11%. Concurrent prescriptions of benzodiazepines and opioids after LBP or LEP diagnosis were associated with increased risk of long-term opioid use.

    View details for DOI 10.1007/s11606-019-05549-8

    View details for PubMedID 31720966

  • Propensity-matched Comparison of Outcomes and Costs After Macroscopic and Microscopic Anterior Cervical Corpectomy Using a National Longitudinal Database. Spine Ho, A. L., Rezaii, P. G., Pendharkar, A. V., Sussman, E. S., Veeravagu, A., Ratliff, J. K., Desai, A. M. 2019; 44 (21): E1281–E1288


    STUDY DESIGN: A retrospective analysis of national longitudinal database.OBJECTIVE: The aim of this study was to examine the outcomes and cost-effectiveness of operating microscope utilization in anterior cervical corpectomy (ACC).SUMMARY OF BACKGROUND DATA: The operating microscope allows for superior visualization and facilitates ACC with less manipulation of tissue and improved decompression of neural elements. However, many groups report no difference in outcomes with increased cost associated with microscope utilization.METHODS: A longitudinal database (MarketScan) was utilized to identify patients undergoing ACC with or without microscope between 2007 and 2016. Propensity matching was performed to normalize differences between the two cohorts. Outcomes and costs were subsequently compared.RESULTS: A total of 11,590 patients were identified for the "macroscopic" group, while 4299 patients were identified for the "microscopic" group. For the propensity-matched analysis, 4298 patients in either cohort were successfully matched according to preoperative characteristics. Hospital length of stay was found to be significantly longer in the macroscopic group than the microscopic group (1.86 nights vs. 1.56 nights, P < 0.0001). Macroscopic ACC patients had an overall higher rate of readmissions [30-day: 4.2% vs. 3.2%, odds ratio (OR) = 0.76 (0.61-0.96), P = 0.0223; 90-day: 7.0% vs. 5.9%, OR = 0.82 (0.69-0.98), P = 0.0223]. Microscopic ACC patients had a higher rate of discharge to home [86.6% vs. 92.5%, OR = 1.91 (1.65-2.21), P < 0.0001] and lower rates of new referrals to pain management [1.0% vs. 0.4%, OR = 0.42 (0.23-0.74), P = 0.0018] compared with macroscopic ACC. Postoperative complication rate was not found to be significantly different between the groups. Finally, total initial admission charges were not significantly different between the macroscopic and microscopic groups (

  • Conventional Versus Stereotactic Image-guided Pedicle Screw Placement During Posterior Lumbar Fusions: A Retrospective Propensity Score-matched Study of a National Longitudinal Database. Spine Pendharkar, A. V., Rezaii, P. G., Ho, A. L., Sussman, E. S., Veeravagu, A., Ratliff, J. K., Desai, A. M. 2019; 44 (21): E1272–E1280


    STUDY DESIGN: Retrospective 1:1 propensity score-matched analysis on a national longitudinal database between 2007 and 2016.OBJECTIVE: The aim of this study was to compare complication rates, revision rates, and payment differences between navigated and conventional posterior lumbar fusion (PLF) procedures with instrumentation.SUMMARY OF BACKGROUND DATA: Stereotactic navigation techniques for spinal instrumentation have been widely demonstrated to improve screw placement accuracies and decrease perforation rates when compared to conventional fluoroscopic and free-hand techniques. However, the clinical utility of navigation for instrumented PLF remains controversial.METHODS: Patients who underwent elective laminectomy and instrumented PLF were stratified into "single level" and "3- to 6-level" cohorts. Navigation and conventional groups within each cohort were balanced using 1:1 propensity score matching, resulting in 1786 navigated and conventional patients in the single-level cohort and 2060 in the 3 to 6 level cohort. Outcomes were compared using bivariate analysis.RESULTS: For the single-level cohort, there were no significant differences in rates of complications, readmissions, revisions, and length of stay between the navigation and conventional groups. For the 3- to 6-level cohort, length of stay was significantly longer in the navigation group (P < 0.0001). Rates of readmissions were, however, greater for the conventional group (30-day: P = 0.0239; 90-day: P = 0.0449). Overall complications were also greater for the conventional group (P = 0.0338), whereas revision rate was not significantly different between the 2 groups. Total payments were significantly greater for the navigation group in both the single level and 3- to 6-level cohorts (P < 0.0001).CONCLUSION: Although use of navigation for 3- to 6-level instrumented PLF was associated with increased length of stay and payments, the concurrent decreased overall complication and readmission rates alluded to its potential clinical utility. However, for single-level instrumented PLF, no differences in outcomes were found between groups, suggesting that the value in navigation may lie in more complex procedures.LEVEL OF EVIDENCE: 3.

    View details for DOI 10.1097/BRS.0000000000003130

    View details for PubMedID 31634303

  • Trends in Anterior Lumbar Interbody Fusion in the United States: A MarketScan Study From 2007 to 2014. Clinical spine surgery Varshneya, K., Medress, Z. A., Jensen, M., Azad, T. D., Rodrigues, A., Stienen, M. N., Desai, A., Ratliff, J. K., Veeravagu, A. 2019


    BACKGROUND: Although the incidence of spinal fusions has increased significantly in the United States over the last quarter century, national trends of anterior lumbar interbody fusion (ALIF) utilization are not known.PURPOSE: The objective of this study was to characterize trends, clinical characteristics, risk factors associated with, and outcomes of ALIF in the United States.STUDY DESIGN: This was an epidemiological study using national administrative data from the MarketScan database.METHODS: Using a large administrative database, we identified adults who underwent ALIF in the United States from 2007 to 2014. The incidence of ALIF was studied longitudinally over time and across geographic regions in the United States. Data related to postoperative complications, length of stay, readmission, and cost were collected.RESULTS: We identified 49,945 patients that underwent ALIF in the United States between 2007 and 2014. The total number of ALIF procedures increased from 3650 in 2007 to 6151 in 2014, accounting for an average increase of 24.07% annually. The Southern United States performed the highest number of ALIFs. The most common conditions treated were degenerative disc disease and spondylolisthesis. Over one third of patients (34.6%) underwent multilevel fusion. The most common complications were iron deficiency anemia, urinary tract infections, and pulmonary complications. Hospital and physician pay increased significantly during the study period.CONCLUSIONS: For the first time in our knowledge, we identified national trends in ALIF utilization, outcomes, and cost using a large administrative database. Our study reaffirms prior work that has demonstrated low rates of complications, mortality, and readmission following ALIF.LEVEL OF EVIDENCE: Level III.

    View details for DOI 10.1097/BSD.0000000000000904

    View details for PubMedID 31609798

  • A Descriptive Analysis of Spinal Cord Arteriovenous Malformations: Clinical Features, Outcomes, and Trends in Management. World neurosurgery Varshneya, K., Pendharkar, A. V., Azad, T. D., Ratliff, J. K., Veeravagu, A. 2019


    BACKGROUND: Spinal arteriovenous malformations (AVM) are an abnormal interconnection of vasculature in the spine than can lead to significant neurological deficit if left untreated.OBJECTIVE: The objective of this study was to characterize how spinal AVM patients initially presented, what treatment options were utilized, and their overall outcomes on a national scale.METHODS: The MarketScan database was queried to identify adult patients diagnosed with a spinal AVM from 2007 - 2015. Trends in management, postoperative complication rates, and costs were determined.RESULTS: 976 patients were identified with having a diagnosis of a spinal AVM. Patients were more commonly treated with an open incision than an embolization (40.1% vs 15.4%). The overall complication rate was 33.61%. Spinal AVM admissions have been stable over the past decade and mean cost of hospitalization has risen from of

Our Team

The Stanford Neurosurgical Ariticial Intelligence and Machine Learning Laboratory is led by Dr. Anand Veeravagu, an Assistant Professor of Neurosurgery and Assistant Professor of Orthopedic Surgery, by courtesy, and Director of Minimally Invasive NeuroSpine Surgery at Stanford. Our laboratory team includes neurosurgery residents, clinical instructors, and medical students. 

We're Hiring!

We are currently looking for post-doctoral fellows looking to build their career in health policy research with specific attention to neurologic diseases. To apply, please contact us by email: