Dr. Hossein Mohamadipanah joined Stanford in March 2018 and his work focuses on artificial intelligence (deep learning).

Current Role at Stanford

Senior Researcher

Education & Certifications

  • Postdoctoral, University of Wisconsin Madison, Data Analysis and Machine Learning (2017)
  • Postdoctoral, Oklahoma State Univesity, Machine Leaning (2015)
  • PhD in Mechanical Engineering, Oklahoma State University, Medical Robotics, Machine Learning, and Computer Vision (2014)
  • MSc in Mechanical Engineering, Sharif University of Technology, Robotics (2010)


Professional Interests

Deep Learning, Machine Learning, Computer Vision, Robotics.

Work Experience

  • Instructor, Oklahoma State University (June 1, 2015 - July 31, 2015)

    Measurements and Instrumentation (MAE 3113)


    stillwater, OK


All Publications

  • Deep kernel recursive least-squares algorithm NONLINEAR DYNAMICS Mohamadipanah, H., Heydari, M., Chowdhary, G. 2021
  • Dynamic Visual Feedback During Junctional Tourniquet Training. The Journal of surgical research Xu, J., Kwan, C., Sunkara, A., Mohamadipanah, H., Bell, K., Tizale, M., Pugh, C. M. 2019; 233: 444–52


    BACKGROUND: This project involved the development and evaluation of a new visual bleeding feedback (VBF) system for tourniquet training. We hypothesized that dynamic VBF during junctional tourniquet training would be helpful and well received by trainees.MATERIALS AND METHODS: We designed the VBF to simulate femoral bleeding. Medical students (n=15) and emergency medical service (EMS) members (n=4) were randomized in a single-blind, crossover study to the VBF or without feedback groups. Poststudy surveys assessing VBF usefulness and recommendations were conducted along with participants' reported confidence using a 7-point Likert scale. Data from the different groups were compared using Wilcoxon signed-rank and rank-sum tests.RESULTS: Participants rated the helpfulness of the VBF highly (6.53/7.00) and indicated they were very likely to recommend the VBF simulator to others (6.80/7.00). Pre- and post-VBF confidence were not statistically different (P=0.59). Likewise, tourniquet application times for VBF and without feedback before crossover were not statistically different (P=0.63). Although participant confidence did not change significantly from beginning to end of the study (P=0.46), application time was significantly reduced (P=0.001).CONCLUSIONS: New tourniquet learners liked our VBF prototype and found it useful. Although confidence did not change over the course of the study for any group, application times improved. Future studies using outcomes of this study will allow us to continue VBF development as well as incorporate other quantitative measures of task performance to elucidate VBF's true benefit and help trainees achieve mastery in junctional tourniquet skills.

    View details for PubMedID 30502284

  • Combining metrics from clinical simulators and sensorimotor tasks can reveal the training background of surgeons. IEEE transactions on bio-medical engineering Huang, F. C., Mohamadipanah, H. n., Mussa-Ivaldi, F. n., Pugh, C. n. 2019


    Skill assessment in surgery traditionally has relied on expert observation and qualitative scoring. Our novel study design demonstrates how analysis of performance in sensorimotor tasks and bench-top surgical simulators can provide inferences about the technical proficiency as well as the training history of surgeons.Our unique study design examined metrics for basic sensorimotor tasks in a virtual reality interface as well as motion metrics in clinical scenario simulations. As indicators of training level, we considered survey responses from surgery residents, including the number of years post-graduation (PGY, four levels), research years (RY, three levels), and clinical years (CY, three levels). Next, we performed a linear discriminant analysis with cross-validation (90% training, 10% testing) to relate the training levels to the selected metrics.Using combined metrics from all stations, we found greater than chance predictions for each survey category, with an overall accuracy of 43.4±2.9% for identifying the level for post-graduate years, 79.1±1.0% accuracy for research training years, and 64.2±1.0% for clinical training years. Our main finding was that combining metrics from all stations resulted in more accurate predictions than using only sensorimotor or clinical scenario tasks. In addition, our analysis indicates that metrics related to the ability to cope with changes in the task environment were the most important predictors of training level.These results suggest that each simulator type provided crucial information for evaluating surgical proficiency. The methods developed in this study could improve evaluations of a surgeon's clinical proficiency as well as training potential in terms of basic sensorimotor ability.

    View details for DOI 10.1109/TBME.2019.2892342

    View details for PubMedID 30629489

  • Shortcut assessment: Can residents' operative performance be determined in the first five minutes of an operative task? Surgery Mohamadipanah, H., Nathwani, J., Peterson, K., Forsyth, K., Maulson, L., DiMarco, S., Pugh, C. 2018; 163 (6): 1207–12


    BACKGROUND: The aim was to validate the potential use of a single, early procedure, operative task as a predictive metric for overall performance. The authors hypothesized that a shortcut psychomotor assessment would be as informative as a total procedural psychomotor assessment when evaluating laparoscopic ventral hernia repair performance on a simulator.METHODS: Using electromagnetic sensors, hand motion data were collected from 38 surgery residents during a simulated laparoscopic ventral hernia repair procedure. Three time-based phases of the procedure were defined: Early Phase (start time through completion of first anchoring suture), Mid Phase (start time through completion of second anchoring suture), and Total Operative Time. Correlations were calculated comparing time and motion metrics for each phase with the final laparoscopic ventral hernia repair score.RESULTS: Analyses revealed that execution time and motion, for the first anchoring suture, predicted procedural outcomes. Greater execution times and path lengths correlated to lesser laparoscopic ventral hernia repair scores (r = -0.56, P = .0008 and r = -0.51, P = .0025, respectively). Greater bimanual dexterity measures correlated to Greater LVH repair scores (r = + 0.47, P = .0058).CONCLUSIONS: This study provides validity evidence for use of a single, early operative task as a shortcut assessment to predict resident performance during a simulated laparoscopic ventral hernia repair procedure. With the continued development and decreasing costs of motion technology, faculty should be well-versed in the use of motion metrics for performance measurements. The results strongly support the use of dexterity and economy of motion (path length + execution time) metrics as early predictors of operative performance.

    View details for PubMedID 29728259

  • Residents' response to bleeding during a simulated robotic surgery task JOURNAL OF SURGICAL RESEARCH Walker, J. L., Nathwani, J. N., Mohamadipanah, H., Laufer, S., Jocewicz, F. F., Gwillim, E., Pugh, C. M. 2017; 220: 385–90
  • Residents' surgical performance during the laboratory years: an analysis of rule-based errors JOURNAL OF SURGICAL RESEARCH Nathwani, J. N., Wise, B. J., Garren, M. E., Mohamadipanah, H., Van Beek, N., DiMarco, S. M., Pugh, C. M. 2017; 219: 226–31


    Nearly one-third of surgical residents will enter into academic development during their surgical residency by dedicating time to a research fellowship for 1-3 y. Major interest lies in understanding how laboratory residents' surgical skills are affected by minimal clinical exposure during academic development. A widely held concern is that the time away from clinical exposure results in surgical skills decay. This study examines the impact of the academic development years on residents' operative performance. We hypothesize that the use of repeated, annual assessments may result in learning even without individual feedback on participants simulated performance.Surgical performance data were collected from laboratory residents (postgraduate years 2-5) during the summers of 2014, 2015, and 2016. Residents had 15 min to complete a shortened, simulated laparoscopic ventral hernia repair procedure. Final hernia repair skins from all participants were scored using a previously validated checklist. An analysis of variance test compared the mean performance scores of repeat participants to those of first time participants.Twenty-seven (37% female) laboratory residents provided 2-year assessment data over the 3-year span of the study. Second time performance revealed improvement from a mean score of 14 (standard error = 1.0) in the first year to 17.2 (SD = 0.9) in the second year, (F[1, 52] = 5.6, P = 0.022). Detailed analysis demonstrated improvement in performance for 3 grading criteria that were considered to be rule-based errors. There was no improvement in operative strategy errors.Analysis of longitudinal performance of laboratory residents shows higher scores for repeat participants in the category of rule-based errors. These findings suggest that laboratory residents can learn from rule-based mistakes when provided with annual performance-based assessments. This benefit was not seen with operative strategy errors and has important implications for using assessments not only for performance analysis but also as a learning experience.

    View details for DOI 10.1016/j.jss.2017.05.129

    View details for Web of Science ID 000413775200034

    View details for PubMedID 29078886

  • Robust Automatic Feature Tracking on Beating Human Hearts for Minimally Invasive CABG Surgery JOURNAL OF MEDICAL DEVICES-TRANSACTIONS OF THE ASME Mohamadipanah, H., Andalibi, M., Hoberock, L. 2016; 10 (4)

    View details for DOI 10.1115/1.4033301

    View details for Web of Science ID 000395330300011

  • Can a virtual reality assessment of fine motor skill predict successful central line insertion? AMERICAN JOURNAL OF SURGERY Mohamadipanah, H., Parthiban, C., Nathwani, J., Rutherford, D., DiMarco, S., Pugh, C. 2016; 212 (4): 573-+


    Due to the increased use of peripherally inserted central catheter lines, central lines are not performed as frequently. The aim of this study is to evaluate whether a virtual reality (VR)-based assessment of fine motor skills can be used as a valid and objective assessment of central line skills.Surgical residents (N = 43) from 7 general surgery programs performed a subclavian central line in a simulated setting. Then, they participated in a force discrimination task in a VR environment. Hand movements from the subclavian central line simulation were tracked by electromagnetic sensors. Gross movements as monitored by the electromagnetic sensors were compared with the fine motor metrics calculated from the force discrimination tasks in the VR environment.Long periods of inactivity (idle time) during needle insertion and lack of smooth movements, as detected by the electromagnetic sensors, showed a significant correlation with poor force discrimination in the VR environment. Also, long periods of needle insertion time correlated to the poor performance in force discrimination in the VR environment.This study shows that force discrimination in a defined VR environment correlates to needle insertion time, idle time, and hand smoothness when performing subclavian central line placement. Fine motor force discrimination may serve as a valid and objective assessment of the skills required for successful needle insertion when placing central lines.

    View details for DOI 10.1016/j.amjsurg.2016.06.023

    View details for Web of Science ID 000385925900003

    View details for PubMedID 27649977

  • Predictive Model Reference Adaptive Controller to Compensate Heart Motion in Minimally Invasive CABG Surgery CARDIOVASCULAR ENGINEERING AND TECHNOLOGY Mohamadipanah, H., Hoberock, L. L., Andalibi, M. 2015; 6 (3): 329–39


    Heart beating is a major challenge in minimally invasive coronary artery surgery. A promising solution is to develop a motion compensation robotic system that gives the surgeon an impression of operating on motionless tissue by synchronizing the surgical tool automatically with the heart tissue motion. To achieve higher control accuracy, an intelligent controller called Predictive Model Reference Adaptive Controller is presented herein, which is adapted not only by observed reference signals but, also by unknown reference signals that are not observed by a camera but could be predicted by a state space estimator. To develop such a system, first the heart surface motion is tracked by the Lucas-Kanade tracking algorithm and validated by human observation. The results of implementing this control algorithm on a real human heart data set show capability of achieving a motion compensation system with high control accuracy.

    View details for DOI 10.1007/s13239-015-0225-y

    View details for Web of Science ID 000380356800011

    View details for PubMedID 26577364

  • Effects of texture addition on optical flow performance in images with poor texture IMAGE AND VISION COMPUTING Andalibi, M., Hoberock, L. L., Mohamadipanah, H. 2015; 40: 1–15
  • Sensors and Psychomotor Metrics: A Unique Opportunity to Close the Gap on Surgical Processes and Outcomes ACS BIOMATERIALS SCIENCE & ENGINEERING Mohamadipanah, H., Perrone, K. H., Peterson, K., Nathwani, J., Huang, F., Garren, A., Garren, M., Witt, A., Pugh, C. 2020; 6 (5): 2630–40
  • Translating motion tracking data into resident feedback: An opportunity for streamlined video coaching Perrone, K. H., Yang, S., Mohamadipanah, H., Wise, B., Witt, A., Goll, C., Pugh, C. EXCERPTA MEDICA INC-ELSEVIER SCIENCE INC. 2020: 552–56


    We hypothesized that differences in motion data during a simulated laparoscopic ventral hernia repair (LVH) can be used to stratify top and lower tier performers and streamline video review.Surgical residents (N = 94) performed a simulated partial LVH repair while wearing motion tracking sensors. We identified the top ten and lower ten performers based on a final product quality score (FPQS) of the repair. Two blinded raters independently reviewed motion plots to identify patterns and stratify top and lower tier performers.Top performers had significantly higher FPQS (23.3 ± 1.2 vs 5.7 ± 1.6 p < 0.01). Raters identified patterns and stratified top performers from lower tier performers (Rater 1 χ2 = 3.2 p = 0.07 and Rater 2 χ2 = 2.0 p = 0.16). During video review, we correlated motion plots with the relevant portion of the procedure.Differences in motion data can identify learning needs and enable rapid review of surgical videos for coaching.

    View details for DOI 10.1016/j.amjsurg.2020.01.032

    View details for Web of Science ID 000525802700004

    View details for PubMedID 32014295

  • Sensors and Psychomotor Metrics: A Unique Opportunity to Close the Gap on Surgical Processes and Outcomes. ACS biomaterials science & engineering Mohamadipanah, H. n., Perrone, K. H., Peterson, K. n., Nathwani, J. n., Huang, F. n., Garren, A. n., Garren, M. n., Witt, A. n., Pugh, C. n. 2020; 6 (5): 2630–40


    The surgical process remains elusive to many. This paper presents two independent empirical investigations where psychomotor skill metrics were used to quantify elements of the surgical process in a procedural context during surgical tasks in a simulated environment. The overarching goal of both investigations was to address the following hypothesis: Basic motion metrics can be used to quantify specific aspects of the surgical process including instrument autonomy, psychomotor efficiency, procedural readiness, and clinical errors. Electromagnetic motion tracking sensors were secured to surgical trainees' (N = 64) hands for both studies, and several motion metrics were investigated as a measure of surgical skill. The first study assessed performance during a bowel repair and laparoscopic ventral hernia (LVH) repair in comparison to a suturing board task. The second study assessed performance in a VR task in comparison to placement of a subclavian central line. The findings of the first study support our subhypothesis that motion metrics have a generalizable application to surgical skill by showing significant correlations in instrument autonomy and psychomotor efficiency during the suturing task and bowel repair (idle time: r = 0.46, p < 0.05; average velocity: r = 0.57, p < 0.05) and the suturing task and LVH repair (jerk magnitude: r = 0.36, p < 0.05; bimanual dexterity: r = 0.35, p < 0.05). In the second study, performance in VR (steering and jerkiness) correlated to clinical errors (r = 0.58, p < 0.05) and insertion time (r = 0.55, p < 0.05) in placement of a subclavian central line. Both gross (dexterity) and fine motor skills (steering) were found to be important as well as efficiency (i.e., idle time, duration, velocity) when seeking to understand the quality of surgical performance. Both studies support our hypotheses that basic motion metrics can be used to quantify specific aspects of the surgical process and that the use of different technologies and metrics are important for comprehensive investigations of surgical skill.

    View details for DOI 10.1021/acsbiomaterials.9b01019

    View details for PubMedID 33463275

  • Multi-Modal Cardiopulmonary Bypass Skills Assessment within a High-Fidelity Simulation Environment. The Annals of thoracic surgery Hermsen, J. L., Mohamadipanah, H. n., Yang, S. n., Wise, B. n., Fiedler, A. n., DiMusto, P. n., Pugh, C. n. 2020


    A high fidelity simulator that utilizes a perfused porcine heart, cannulae, and tubing has demonstrated to be a useful training adjunct. We hypothesized that multi-modal assessment of cardiopulmonary bypass skills within this high fidelity simulated environment could discern expert from trainee performance.Three traditional fellows (PGY 6-8) and three attending surgeons each performed three aortic cannulations. The third sequence included venous cannulation, commencement of cardiopulmonary bypass (CPB) and placement of a cardioplegia catheter and aortic crossclamp (XC). Performance across 20 cognitive and 21 technical domains were evaluated. Surgeon and assistant hand movements and economy of motion were assessed by electromagnetic motion sensors worn under sterile gloves.Analysis showed a significant difference in cognitive (6.7±2.3 vs 4.6±2.7, p=0.03) but not technical (6.2±2.5 vs 5.8±2.2, p=0.7) scores favoring the experts. In addition, experts showed higher efficiency by spending 64±14s to construct a non-pledgeted aortic pursestring suture and secure it with a Rummel while trainees spent 82±30s to complete this task (p=0.03). Motion analysis revealed non-different path lengths between experts and trainees for cannulation and CPB, but significantly shorter path for experts in XC (47.5±15.5m vs 91.9±20.3m, p=0.04).Multi-modal assessment using cognitive, technical and motion analysis of basic cardiopulmonary bypass tasks using a high-fidelity simulation environment is a valid system to measure performance and discriminate experts from trainees. This construct may allow for development of 'competence thresholds' with important implications for training and certification in cardiothoracic surgery.

    View details for DOI 10.1016/j.athoracsur.2020.07.022

    View details for PubMedID 32971063

  • Use of sensors to quantify procedural idle time: Validity evidence for a new mastery metric. Surgery Perrone, K. H., Yang, S., Wise, B., Witt, A., Goll, C., Dawn, S., Eichhorn, W., Mohamadipanah, H., Pugh, C. 2019


    BACKGROUND: Quantification of mastery is the first step in using objective metrics for teaching. We hypothesized that during orotracheal intubation, top tier performers have less idle time compared to lower tier performers.METHODS: At the Anesthesiology 2018 Annual Meeting, 82 participants intubated a normal airway simulator and a burnt airway simulator. The movements of the participant's laryngoscope were quantified using electromagnetic motion sensors. Top tier performers were defined as participants who intubated both simulators successfully in less than the median time for each simulator. Idle time was defined as the duration of time when the laryngoscope was not moving.RESULTS: Top performers showed less Idle Time when intubating the normal airway compared to lower tier performers (14.5 ± 9.8 seconds vs 34.0 ± 52.0 seconds, respectively P < .01). Likewise, top performers showed less Idle Time when intubating the burnt airway compared to lower tier performers (18.6 ± 15.2 seconds vs 63.4 ± 59.11 seconds; P < .01). Comparing performance on the burnt airway to the normal airway, there was a difference for lower tier performers (63.4 ± 59.1 seconds vs 34.0 ± 52.0 seconds; P < .01) but not for top tier performers (18.6 ± 15.2 seconds vs 14.5 ± 9.8 seconds; P= .07).CONCLUSION: Similar to our previous findings with other procedures, Idle Time was shown to have known group validity evidence when comparing top performers with lower tier performers. Further, Idle Time was correlated with procedure difficulty in our prior work. We observed statistically significant differences in Idle Times for lower tier performers when comparing the normal airway to the burnt airway but not for top tier performers. Our findings support the continued exploration of Idle Time for development of objective assessment and curricula.

    View details for DOI 10.1016/j.surg.2019.09.016

    View details for PubMedID 31708084

  • Screening surgical residents' laparoscopic skills using virtual realitytasks: Who needs more time in the sim lab? Surgery Mohamadipanah, H., Perrone, K. H., Nathwani, J., Parthiban, C., Peterson, K., Wise, B., Garren, A., Pugh, C. 2019


    BACKGROUND: This study investigated the possibility of using virtual reality perceptual-motor tasks as a screening tool for laparoscopic ability. We hypothesized that perceptual-motor skills assessed using virtual reality will correlate with the quality of simulated laparoscopic ventral hernia repair.MATERIALS AND METHODS: Surgical residents (N= 37), performed 2 virtual reality perceptual-motor tasks: (1) force matching and (2) target tracking. Participants also performed a laparoscopic ventral hernia repair on a simulator and final product quality score, and endoscopic visualization errors were calculated. Correlational analysis was performed to assess the relationship between performance on virtual reality tasks and laparoscopic ventral hernia repair.RESULTS: Residents with poor performance on force matching in virtual reality-"peak deflection" (r= -0.34, P < .05) and "summation distance" (r= -0.36, P < .05)-had lower final product quality scores. Likewise, poor performance in virtual reality-based target tracking-"path length" (r= -0.49, P < .05) and "maximum distance" (r= -0.37, P < .05)-correlated with a lower final product quality score.CONCLUSION: Our findings support the notion that virtual reality could be used as a screening tool for perceptual-motor skill. Trainees identified as having poor perceptual-motor skill can benefit from focused curricula, allowing them to hone personal areas of weakness and maximize technical skill.

    View details for DOI 10.1016/j.surg.2019.04.013

    View details for PubMedID 31229312

  • Can VR Be Used to Track Skills Decay During the Research Years? The Journal of surgical research Mohamadipanah, H. n., Perrone, K. n., Peterson, K. n., Garren, M. n., Parthiban, C. n., Sunkara, A. n., Zinn, M. n., Pugh, C. n. 2019


    Time away from surgical practice can lead to skills decay. Research residents are thought to be prone to skills decay, given their limited experience and reduced exposure to clinical activities during their research training years. This study takes a cross-sectional approach to assess differences in residents' skills at the beginning and end of their research years using virtual reality. We hypothesized that research residents will have measurable decay in psychomotor skills when evaluated using virtual reality.Surgical residents (n = 28) were divided into two groups; the first group was just beginning their research time (clinical residents: n = 19) and the second group (research residents: n = 9) had just finished at least 2 y of research. All participants were asked to perform a target-tracking task using a haptic device, and their performance was compared using Welch's t-test.Research residents showed a higher level of "tracking error" (1.69 ± 0.44 cm versus 1.40 ± 0.19 cm; P = 0.04) and a similar level of "path length" (62.5 ± 10.5 cm versus 62.1 ± 5.2 cm; P = 0.92) when compared with clinical residents.The increased "tracking error" among residents at the end of their research time suggests fine psychomotor skills decay in residents who spend time away from clinical duties during laboratory time. This decay demonstrates the need for research residents to regularly participate in clinical activities, simulation, or assessments to minimize and monitor skills decay while away from clinical practice. Additional longitudinal studies may help better map learning and decay curves for residents who spend time away from clinical practice.

    View details for DOI 10.1016/j.jss.2019.10.030

    View details for PubMedID 31776024

Latest information on COVID-19