All Publications

  • Perspectives on validation of clinical predictive algorithms. NPJ digital medicine de Hond, A. A., Shah, V. B., Kant, I. M., Van Calster, B., Steyerberg, E. W., Hernandez-Boussard, T. 2023; 6 (1): 86

    View details for DOI 10.1038/s41746-023-00832-9

    View details for PubMedID 37149704

  • Clinical outcomes and cost differences between patients undergoing primary anterior cervical discectomy and fusion procedures with private or Medicare insurance: a propensity score matched study. World neurosurgery Shah, V., Rodrigues, A. J., Malhotra, S., Johnstone, T., Varshneya, K., Haider, G., Stienen, M. N., Veeravagu, A. 2023


    To assess whether insurance type reflects a patient's quality of care following an ACDF procedure, by comparing differences in post-operative complications, readmission rates, reoperation rates, length of hospital stay, and cost of treatment between patients with Medicare versus private insurance.Propensity score matching (PSM) was employed to match patient cohorts insured by Medicare and private insurance in the MarketScan Commercial Claims and Encounters Database (2007-2016). Age, sex, year of operation, geographic region, comorbidities, and operative factors were used to match cohorts of patients undergoing an ACDF procedure.A total of 110,911 patients met the inclusion criteria, of which 97,543 patients (87.9%) were privately insured and 13,368 patients (12.1%) were insured by Medicare. The PSM algorithm matched 7,026 privately insured patients to 7,026 Medicare patients. After matching, there was no significant difference in 90-day post-operative complication rates, length of stay, or reoperation rates between the Medicare and privately insured cohorts. The Medicare group had lower post-operative readmission rates for all time points: 30 days (1.8% vs. 4.6%; p < 0.001), 60 days (2.5% vs. 6.3%; p < 0.001), and 90 days (4.2% vs. 7.7%; p < 0.001). The median payments to physicians were significantly lower for the Medicare group ($3,885 vs. $5,601; p < 0.001).In this study, propensity score matched patients covered by Medicare and private insurance that underwent an ACDF procedure were found to have similar treatment outcomes.

    View details for DOI 10.1016/j.wneu.2023.02.129

    View details for PubMedID 36871653

  • Polyvinylpyrrolidone-Coated Catheters Decrease Astrocyte Adhesion and Improve Flow/Pressure Performance in an Invitro Model of Hydrocephalus. Children (Basel, Switzerland) Castaneyra-Ruiz, L., Lee, S., Chan, A. Y., Shah, V., Romero, B., Ledbetter, J., Muhonen, M. 2022; 10 (1)


    The leading cause of ventricular shunt failure in pediatric patients is proximal catheter occlusion. Here, we evaluate various types of shunt catheters to assess in vitro cellular adhesion and obstruction. The following four types of catheters were tested: (1) antibiotic- and barium-impregnated, (2) polyvinylpyrrolidone, (3) barium stripe, and (4) barium impregnated. Catheters were either seeded superficially with astrocyte cells to test cellular adhesion or inoculated with cultured astrocytes into the catheters to test catheter performance under obstruction conditions. Ventricular catheters were placed into a three-dimensional printed phantom ventricular replicating system through which artificial CSF was pumped. Differential pressure sensors were used to measure catheter performance. Polyvinylpyrrolidone catheters had the lowest median cell attachment compared to antibiotic-impregnated (18 cells), barium stripe (17 cells), and barium-impregnated (21.5 cells) catheters after culture (p < 0.01). In addition, polyvinylpyrrolidone catheters had significantly higher flow in the phantom ventricular system (0.12 mL/min) compared to the antibiotic coated (0.10 mL/min), barium stripe (0.02 mL/min) and barium-impregnated (0.08 mL/min; p < 0.01) catheters. Polyvinylpyrrolidone catheters showed less cellular adhesion and were least likely to be occluded by astrocyte cells. Our findings can help suggest patient-appropriate proximal ventricular catheters for clinical use.

    View details for DOI 10.3390/children10010018

    View details for PubMedID 36670569

  • Tutorial: implementing and visualizing machine learning (ML) clinical prediction models into web-accessible calculators using Shiny R ANNALS OF TRANSLATIONAL MEDICINE Eddington, H. S., Trickey, A. W., Shah, V., Harris, A. S. 2022
  • Inspiration out of Trauma: Meet Resurge Exchange Scholar Dr. David Silver Shah, V. ReSurge International. 2022
  • Meet the First Annual Matolase Mtonga Scholarship Winners Shah, V. ReSurge International. 2022
  • Algorithmic Prediction of Delayed Radiology Turn-Around-Time during Non-Business Hours. Academic radiology Shah, V., Chillakuru, Y. R., Rybkin, A., Seo, Y., Vu, T., Sohn, J. H. 2021


    Radiology turnaround time is an important quality measure that can impact hospital workflow and patient outcomes. We aimed to develop a machine learning model to predict delayed turnaround time during non-business hours and identify factors that contribute to this delay.This retrospective study consisted of 15,117 CT cases from May 2018 to May 2019 during non-business hours at two hospital campuses after applying exclusion criteria. Of these 15,177 cases, 7,532 were inpatient cases and 7,585 were emergency cases. Order time, scan time, first communication by radiologist, free-text indications, and other clinical metadata were extracted. A combined XGBoost classifier and Random Forest natural language processing model was trained with 85% of the data and tested with 15% of the data. The model predicted two measures of delay: when the exam was ordered to first communication (total time) and when the scan was completed to first communication (interpretation time). The model was analyzed with the area under the curve (AUC) of receiver operating characteristic (ROC) and feature importance. Source code: RESULTS: The algorithm reached an AUC of 0.85, with a 95% confidence interval [0.83, 0.87], when predicting delays greater than 245 minutes for "total time" and 0.71, with a 95% confidence interval [0.68, 0.73], when predicting delays greater than 57 minutes for "interpretation time". At our institution, CT scan description (e.g. "CTA chest pulmonary embolism protocol"), time of day, and year in training were more predictive features compared to body part, inpatient status, and hospital campus for both interpretation and total time delay.This algorithm can be applied clinically when a physician is ordering the scan to reasonably predict delayed turnaround time. Such a model can be leveraged to identify factors associated with delays and emphasize areas for improvement to patient outcomes.

    View details for DOI 10.1016/j.acra.2021.05.026

    View details for PubMedID 34187741

  • I Can't Breathe, Pollution, and Racial Justice Shah, V. Society for Social Studies of Science. 4S Backchannels. 2021
  • Intrathecal Access Through Suboccipital Port in Patients With Spinal Muscular Atrophy and Complex Spines: Case Series and Technical Note. Cureus Tran, D. K., Shah, V., Muhonen, M. G. 2020; 12 (8): e9525


    Introduction Many patients with the spinal muscular atrophy (SMA) have complex spinal anatomy, secondary to thoraco-lumbar spinal fusions. Their fragile musculoskeletal anatomy potentiates limb and joint injury if conventional spinal fluid access modalities are utilized. This creates a challenge when attempting to deliver intrathecal medications such as nusinersen (Spinraza®). Catheter placement in the cervical subarachnoid space with a caudally directed tip is potentially beneficial. This article describes our experience with Spinraza injections into the thecal space through a suboccipital port. This allowed for simple, chronic, and reliable cerebrospinal fluid (CSF) aspiration and intrathecal injections. Methods A total of 15 patients with SMA and complex spinal anatomy were implanted with a cervical subarachnoid catheter, connected to a suboccipital access port. We retrospectively reviewed the charts of these patients for clinical outcomes and complications. All patients then underwent serial port cannulation, aspiration of CSF, and injection of Spinraza following standard manufacturer dosage guidelines. Results The age range was 3 to 49. Two had type-1 SMA, 10 had type-2 SMA, and three had type-3 SMA. We were able to successfully cannulate the port, aspirate CSF, and inject Spinraza during all access attempts. Two incidents of subcutaneous CSF leaks were resolved through reoperation and one incident of transient CSF leak was resolved without surgical repair. Conclusion Patients with SMA requiring intrathecal injections of Spinraza can be treated safely and efficiently with this novel implantation technique. The complication rates are low and the injection time is dramatically lower than with conventional injection techniques.

    View details for DOI 10.7759/cureus.9525

    View details for PubMedID 32905153

    View details for PubMedCentralID PMC7465927