MIMicah Grubert Van Iderstine

Micah Grubert Van Iderstine

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Micah Grubert Van Iderstine has a total of 39 co-authors. Here are the top co-authors by number of shared publications:

  • Gerald Yosel Minuk

    Departments of Internal Medicine, University of Manitoba, Winnipeg, Manitoba, Canada

    4 shared publications
  • Julia Uhanova

    Department of Internal Medicine, University of Manitoba Max Rady College of Medicine, Winnipeg, Manitoba, Canada

    3 shared publications
  • Nabiha Faisal

    Section of Hepatology, Department of Internal Medicine, Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada

    3 shared publications
  • Caelan Stephanson

    Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada

    2 shared publications
  • Megan Weisshaar

    Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada

    2 shared publications
  • Sem Perez

    Max Rady College of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada

    2 shared publications

Latest Publications

    • 1 April 2026
      • Vol. 46  (4) , e70560
    Open Access

    Early Prediction of Hepatic Decompensation in Cirrhosis Using Optimised XGBoost Models at the Initial Outpatient Hepatology Visit.

    by Micah Grubert Van Iderstine, Sem Perez, Gregory S Jackson, Tyler Szun, Caelan Stephanson, Megan Weisshaar, Gerald Y Minuk, Nabiha Faisal

    Hepatic decompensation represents a critical transition in cirrhosis, leading to increased morbidity, mortality and healthcare utilisation. Identifying patients at risk of decompensation remains a clinical challenge. We aimed to develop and validate XGBoost models to predict hepatic decompensation at multiple time points using clinical data available at a patient's initial outpatient hepatology visit. We conducted a retrospective cohort study including 2208 adult patients with cirrhosis or its complications seen in hepatology clinics between 1985 and 2022. Patients were classified as compensated or decompensated based on a keyword search of the Philip and Ellie Kives Clinical Database, with decompensation dates confirmed by chart review. Sixteen routinely available variables including demographics, biochemical parameters and disease aetiology were used as predictors. Logistic regression and XGBoost models were trained to predict hepatic decompensation at 1, 3, 5 and 10 years, with a random 20% holdout test set used for validation. XGBoost models were tuned to optimise the precision-recall area under the curve (PR-AUC). Performance was evaluated using AUROC, precision, recall and F1 scores. XGBoost models outperformed logistic regression at most time points, demonstrating strong performance at 3 and 5 years. Recall was 0.42, 0.98, 0.98 and 0.82 at 1, 3, 5 and 10 years respectively. Corresponding AUROC values were 0.85, 0.88, 0.81 and 0.89. Optimised XGBoost models demonstrated robust predictive accuracy for medium- and long-term hepatic decompensation among patients with compensated cirrhosis. These models may support early risk stratification and enable personalised management strategies to prevent clinical deterioration.

  • Open Access

    Early Prediction of Hepatic Decompensation in Cirrhosis Using Optimised XGBoost Models at the Initial Outpatient Hepatology Visit

    by Micah Grubert Van Iderstine, Sem Perez, Gregory S. Jackson, Tyler Szun, Caelan Stephanson, Megan Weisshaar, Gerald Y. Minuk, Nabiha Faisal

    Background and Aims: Hepatic decompensation represents a critical transition in cirrhosis, leading to increased morbidity, mortality and healthcare utilisation. Identifying patients at risk of decompensation remains a clinical challenge. We aimed to develop and validate XGBoost models to predict hepatic decompensation at multiple time points using clinical data available at a patient's initial outpatient hepatology visit. Methods: We conducted a retrospective cohort study including 2208 adult patients with cirrhosis or its complications seen in hepatology clinics between 1985 and 2022. Patients were classified as compensated or decompensated based on a keyword search of the Philip and Ellie Kives Clinical Database, with decompensation dates confirmed by chart review. Sixteen routinely available variables including demographics, biochemical parameters and disease aetiology were used as predictors. Logistic regression and XGBoost models were trained to predict hepatic decompensation at 1, 3, 5 and 10 years, with a random 20% holdout test set used for validation. XGBoost models were tuned to optimise the precision‐recall area under the curve (PR‐AUC). Performance was evaluated using AUROC, precision, recall and F1 scores. Results: XGBoost models outperformed logistic regression at most time points, demonstrating strong performance at 3 and 5 years. Recall was 0.42, 0.98, 0.98 and 0.82 at 1, 3, 5 and 10 years respectively. Corresponding AUROC values were 0.85, 0.88, 0.81 and 0.89. Conclusions: Optimised XGBoost models demonstrated robust predictive accuracy for medium‐ and long‐term hepatic decompensation among patients with compensated cirrhosis. These models may support early risk stratification and enable personalised management strategies to prevent clinical deterioration.


  • Open Access

    A randomized controlled study on medical students learning anatomy through hands‐on ultrasound

    by Micah Grubert Van Iderstine, Jacob Jensen, Tomislav Jelic, Terry Y. Li

    Anatomy education in the undergraduate medical curriculum faces many competing interests, including the increasing demand for ultrasound learning. Instead of being treated as a separate subject, ultrasound can offer a unique lens to visualize anatomy without using cadaveric materials. This study used a randomized controlled design to compare learning with hands‐on ultrasound to learning on cadavers. Forty preclinical medical students were randomized to an experimental group or a control group. Both groups attended a session covering structures in the neck, upper limb, and abdomen. The experimental group learned using hands‐on ultrasound imaging, while the control group was taught on a dissected cadaver. Participants completed multiple‐choice tests on anatomical relationships and structure identification at four time points: presession, post‐session, 1‐week follow‐up, and 1‐month follow‐up. There was no statistical difference between test scores of the two groups. The post‐session average score (54% for cadaver group, 57% for ultrasound group) more than doubled the presession average score (20% for cadaver group, 25% for ultrasound group) (p < 0.001). One‐week follow‐up and 1‐month follow‐up scores (40%–44%) significantly decreased from immediate post‐session for both groups. Eighteen of the 20 ultrasound‐facilitated participants felt more confident operating an ultrasound device compared to before the session. Three quarters of all participants agreed hands‐on ultrasound should be part of their anatomy education. This study shows hands‐on ultrasound imaging can effectively substitute cadaver‐based learning of certain anatomy topics at the introductory level. Ultrasound integration highlights the clinical relevance of anatomy and provides an innovative tool for anatomical education.


  • Open Access

    Improving prognostic accuracy in lung transplantation using unique features of isolated human lung radiographs

    by Bonnie T. Chao, Andrew T. Sage, Micheal C. McInnis, Jun Ma, Micah Grubert Van Iderstine, Xuanzi Zhou, Jerome Valero, Marcelo Cypel, Mingyao Liu, Bo Wang, et al.

    Ex vivo lung perfusion (EVLP) enables advanced assessment of human lungs for transplant suitability. We developed a convolutional neural network (CNN)-based approach to analyze the largest cohort of isolated lung radiographs to date. CNNs were trained to process 1300 longitudinal radiographs from n = 650 clinical EVLP cases. Latent features were transformed into principal components (PC) and correlated with known radiographic findings. PCs were combined with physiological data to classify clinical outcomes: (1) recipient time to extubation of p < 0·0001), and adding radiographic PCs significantly improved the discrimination for clinical outcomes (Accuracy: 73 vs 78%, p = 0·014). CNN-derived radiographic lung features therefore add substantial value to the current assessments. This approach can be adopted by EVLP centers worldwide to harness radiographic information without requiring real-time radiological expertise.


  • Comparison of different definitions of metabolic syndrome and their associations with non-alcoholic fatty liver disease: a retrospective study

    by Carmen Tse, Nicholas Lisanti, Micah Grubert Van Iderstine, Julia Uhanova, Gerald Minuk, Nabiha Faisal

    Background: Metabolic syndrome (MetS) is considered an important risk factor for non-alcoholic fatty liver disease (NAFLD). The aim of this study was to measure the prevalence of MetS based on six different MetS definitions and compare the performance of various definitions for identifying diabetes, hypertension, and dyslipidemia among NAFLD patients. Methods: The definitions compared were those developed by the World Health Organization (WHO), National Cholesterol Education Program Adult Treatment Panel III (NCEP-ATP III), International Diabetes Federation (IDF), American Association of Clinical Endocrinologists (AACE), American Heart Association/National Heart, Lung and Blood Institute (AHA/NHLBI), and Interim Joint Statement “harmonized” criteria. Receiver operator characteristic (ROC) curves were plotted for the six MetS definitions with NAFLD diagnosis. The diagnosis for NAFLD was established based on liver imaging or biopsy compatible with fatty liver disease. Results: A total of 500 NAFLD patients were analyzed. The mean age was 61.2 (SD 13.2) years, and BMI was 32.7 (SD 8.0) kg/m2. The most prevalent MetS component was dyslipidemia (83%), followed by hypertension (60%), obesity (61%), and diabetes (57%). The prevalence of MetS according to the WHO, NCEP/ATP-III, IDF, AACE, AHA/NHLBI, and harmonized criteria was 69%, 59%, 54%, 64%, 78%, and 79%, respectively. The highest area under the ROC curve for diabetes and hypertension was with the WHO definition (0.7405) and (0.8120), respectively. Conclusions: The prevalence of MetS in NAFLD patients varies according to the definitions of MetS employed. The modified WHO definition appeared to be most useful for the screening of MetS in NAFLD patients.


  • Open Access

    The impact of primary biliary cholangitis on non-alcoholic fatty liver disease

    by Daniel Iluz-Freundlich, Julia Uhanova, Micah Grubert Van Iderstine, Gerald Y. Minuk

    Background: The impact of chronic cholestatic liver diseases such as primary biliary cholangitis (PBC) on non-alcoholic fatty liver disease (NAFLD) has yet to be described. Objectives: To document and compare the severity and course of liver disease in patients with NAFLD/PBC versus NAFLD alone. Methods: In this retrospective, case-control study 68 adult NAFLD/PBC patients were matched 1:2 for age and sex with 136 NAFLD alone patients. Disease activity and severity were documented by serum aminotransferases, albumin, bilirubin and international normalized ratio (INR) values and hepatic fibrosis by Fib-4 and aspartate aminotransferase/platelet ratio indices (APRI). Results: On presentation (baseline), NAFLD/PBC patients had similar serum aminotransferase, albumin and bilirubin levels but lower INR values than NAFLD alone patients. Fib-4 and APRI levels were similar. Despite longer follow-up (favouring more advanced disease) in NAFLD/PBC patients, serum aminotransferases and bilirubin values were similar but albumin and INR levels significantly lower in NAFLD/PBC versus NAFLD alone patients at the end of follow-up. NAFLD/PBC patients also had significantly lower and less worsening of Fib-4 values at the end of follow-up. Transition from intermediate Fib-4 levels to those compatible with no or limited fibrosis was higher in NAFLD/PBC patients. Conclusion: These findings suggest PBC does not adversely affect the severity or course of NAFLD.


  • Open Access

    The effects of hyperthermia on human hepatocellular carcinoma stem and mature cancer cells

    by David Sontag, David A. Miles, Julia Uhanova, Micah Grubert Van Iderstine, Jiaqi Yang, Gerald Y. Minuk


  • Does SEN-V and other non-A-E hepatotropic viruses contribute to the development and progression of non-alcoholic fatty liver disease?

    by M. Grubert Van Iderstine, C. Osiowy, C. Rumbolt, M. Zhang, K. Swidinsky, K.D.E. Kaita, G.Y. Minuk


  • Open Access

    Liver disease referrals to an urban, hospital‐based hepatology outpatient clinic over the past 25 years

    by Micah Grubert Van Iderstine, Daniel Iluz‐Freundlich, Casandra Dolovich, Eurielle Villarin, Gerald Y Minuk

    Background: Additional hepatologists are required to manage the rapidly increasing number of patients with liver disease. One disincentive to trainees considering a career in hepatology is the longstanding perception that outpatient hepatology consists largely of managing patients with alcohol‐induced liver disease (ALD).Objectives: To document the types of liver diseases and changes in liver disease referrals to an urban outpatient liver disease clinic over the past 25 years.Methods: The nature of the liver disorder, age, gender, and socioeconomic status of patients referred to an urban, hospital‐based, liver diseases outpatient program were documented from 1992 to 2017. Joinpoint analysis was performed to identify significant trends in referral prevalence rates of various disorders.Results: In 1992/1993, hepatitis C virus (HCV), followed by hepatitis B virus (HBV), “other”, non‐alcoholic fatty liver disease (NAFLD), and primary biliary cholangitis (PBC) were the most common underlying liver diseases in referred patients (39, 36, 12, 4.5, and 3.5% respectively), whereas in 2016/2017, NAFLD, HBV, HCV, “other,” and ALD were most common (60, 15, 12, 8.7, and 3.3%, respectively). Aside from NAFLD referrals, which consistently increased over the 25‐year period, the prevalence of all other liver disease referrals fluctuated but generally declined. Recently referred patients were significantly older (38 ± 13 years in 1992/1993 and 49 ± 15 years in 2016/2017, P < 0.0001), while gender and socioeconomic status have not changed.Conclusions: Hepatology is a diverse, dynamic subspecialty where ALD continues to constitute less than 5% of all patient referrals.