Introduction: This study aimed to train machine learning algorithms (MLAs) to detect advanced fibrosis (AF) in metabolic dysfunction-associated steatotic liver disease (MASLD) patients at the level of primary care setting and to explain the predictions to ensure responsible use by clinicians. Methods: Readily available features of 618 MASLD patients followed up at a tertiary center were used to train five MLAs. AF was defined as liver stiffness ≥9.3 kPa, measured via 2-dimension shear wave elastography (n = 495) or liver biopsy ≥F3 (n = 123). MLAs were compared to Fibrosis-4 index (FIB-4) and non-alcoholic fatty liver disease (NAFLD) fibrosis score (NFS) on 540 MASLD patients from the primary care setting as validation. Feature importance, partial dependence, and shapely additive explanations (SHAPs) were utilized for explanation. Results: Extreme gradient boosting (XGBoost) achieved an AUC = 0.91, outperforming FIB-4 (AUC = 0.78) and NFS (AUC = 0.81, both p < 0.05) with specificity = 76% versus 59% and 48% for FIB-4 ≥1.3 and NFS ≥−1.45, respectively (p < 0.05). Its sensitivity (91%) was superior to FIB-4 (79%). XGBoost confidently excluded AF (negative predictive value = 99%) with the highest positive predictive value (31%), superior to FIB-4 and NFS (all p < 0.05). The most important features were HbA1c and gamma glutamyl transpeptidase (GGT) with a steep increase in AF probability at HbA1c >6.5%. The strongest interaction was between AST and age. XGBoost, but not logistic regression, extracted informative patterns from ALT, low-density lipoprotein cholesterol, and alkaline phosphatase (p < 0.001). One-quarter of the false positives (FPs) were correctly reclassified with only one additional false negative based on the SHAP values of GGT, platelets, and ALT which were found to be associated with a FP classification. Conclusions: An explainable XGBoost algorithm was demonstrated superior to FIB-4 and NFS for screening of AF in MASLD patients at the primary care setting. The algorithm also proved safe for use as clinicians can understand the predictions and flag FP classifications.

Background: The ongoing outbreak of COVID-19 is associated with higher levels of morbidity and mortality among patients with comorbidities, including the metabolic syndrome. Liver impairment has been reported in up to 54% of hospitalized patients with COVID-19. The impact of COVID-19 on a preexisting chronic liver disease is an actively studied area of research. The contribution of our study is towards determining the predictors of severity and the outcome of liver injury among hospitalized patients with COVID-19 infection, including patients with a preexisting liver disease and COVID-19. Methods: This single center retrospective cohort study included all patients ≥18 years, admitted in Sheba Medical Center with confirmed COVID-19 infection. Demographic, clinical and laboratory data were obtained using the MDClone platform and rechecked after data decryption using electronic health records. Results: Of 382 patients with COVID-19, 66.4% had increased liver biochemistry. Mild increase was observed in 76.7%. The higher level of fibrosis-4 (FIB-4) at admission was independently associated with higher mortality rate. Preexisting liver disease was detected in 15.4% patients. Most common etiology was nonalcoholic fatty liver disease (78.7%). The mortality of hospitalized patients with preexisting liver disease was 16.7% compared to 6.8% in patients without preexisting liver disease (RR = 2.792, P = 0.01). In multivariate analysis, liver disease adjusted to age and BMI was associated with mortality with high statistical significance. Conclusions: Patients with preexisting chronic liver disease were at a higher risk of mortality. The FIB-4 level at admission was associated with worse prognosis. These findings should be reevaluated in a larger cohort of patients.

Sinusoidal Obstructive Syndrome (SOS) is a life-threatening complication after hematopoietic stem-cell transplantation (HSCT), characterized by post-sinusoidal portal hypertension. FibroScan is used to assess portal hypertension non-invasively. We assessed transient elastography (TE) applicability in diagnosing SOS. The study included 27 adult patients, 11 underwent TE for high SOS risk pre-HSCT, 17 underwent TE post-HSCT due to bilirubin ≥2 mg/dl with no definite diagnosis of SOS. The first group had median Liver Stiffness Measurement (LSM) of 7.4 kPa (range, 3.3–22.5). Based on LSM results, conditioning regimen was modified for six patients and two of them developed SOS. Only one patient who did not have protocol adjustment experienced SOS. No patient with LSM < 7 kPa developed SOS. The second group had median LSM of 7.7 kPa (4.4–31.5). Median LSM after HSCT was significantly higher in patients who subsequently developed established SOS (n = 10) compared to patients who did not (n = 8), with values of 10.7 kPa (5.6–31.5) and 5.9 kPa (4.4–13.8), respectively (p = 0.02). An LSM cut-off of 7.5 kPa had a sensitivity and specificity of 75 and 80% for diagnosing SOS. In conclusion, pre-HSCT LSM can help adjustment of conditioning regimen in patients with high-risk for SOS. Post-HSCT LSM can help in early diagnosis of SOS.