Artificial intelligence in the diagnosis of cirrhosis and portal hypertension

The Prediction and Treatment of Bleeding Esophageal Varices in the Artificial Intelligence Era: A Review

Esophageal varices (EVs), a significant complication of cirrhosis, present a considerable challenge in clinical practice due to their high risk of bleeding and associated morbidity and mortality. This manuscript explores the transformative role of artificial intelligence (AI) in the management of EV, particularly in enhancing diagnostic accuracy and predicting bleeding risks. It underscores the potential of AI in offering noninvasive, efficient alternatives to traditional diagnostic methods such as esophagogastroduodenoscopy (EGD). The complexity of EV management is highlighted, necessitating a multidisciplinary approach that includes pharmacological therapy, endoscopic interventions, and, in some cases, surgical options tailored to individual patient profiles. Additionally, the paper emphasizes the importance of integrating AI into medical education and practice, preparing healthcare professionals for the evolving landscape of medical technology. It projects a future where AI significantly influences the management of gastrointestinal bleeding, improving clinical decision-making, patient outcomes, and overall healthcare efficiency. The study advocates for a patient-centered approach in healthcare, balancing the incorporation of innovative technologies with ethical principles and the diverse needs of patients to optimize treatment efficacy and enhance healthcare accessibility.

Deep Neural Network Regression to Assist Non-Invasive Diagnosis of Portal Hypertension

Portal hypertension is a complex medical condition characterized by elevated blood pressure in the portal venous system. The conventional diagnosis of such disease often involves invasive procedures such as liver biopsy, endoscopy, or imaging techniques with contrast agents, which can be uncomfortable for patients and carry inherent risks. This study presents a deep neural network method in support of the non-invasive diagnosis of portal hypertension in patients with chronic liver diseases. The proposed method utilizes readily available clinical data, thus eliminating the need for invasive procedures. A dataset composed of standard laboratory parameters is used to train and validate the deep neural network regressor. The experimental results exhibit reasonable performance in distinguishing patients with portal hypertension from healthy individuals. Such performances may be improved by using larger datasets of high quality. These findings suggest that deep neural networks can serve as useful auxiliary diagnostic tools, aiding healthcare professionals in making timely and accurate decisions for patients suspected of having portal hypertension.

Endoscopic Ruler for varix size measurement: A multicenter pilot study

BACKGROUND

We invented Endoscopic Ruler, a new endoscopic device to measure the size of varices in patients with cirrhosis and portal hypertension.

AIM

To assess the feasibility and safety of Endoscopic Ruler, and evaluate the agreement on identifying large oesophageal varices (OV) between Endoscopic Ruler and the endoscopists, as well as the interobserver agreement on diagnosing large OV using Endoscopic Ruler.

METHODS

We prospectively and consecutively enrolled patients with cirrhosis from 11 hospitals, all of whom got esophagogastroduodenoscopy (EGD) with Endoscopic Ruler. The primary study outcome was a successful measurement of the size of varices using Endoscopic Ruler. The secondary outcomes included adverse events, operation time, the agreement of identifying large OV between the objective measurement of Endoscopic Ruler and the empirical reading of endoscopists, together with the interobserver agreement on diagnosing large OV by Endoscopic Ruler.

RESULTS

From November 2020 to April 2022, a total of 120 eligible patients with cirrhosis were recruited and all of them underwent EGD examinations with Endoscopic Ruler successfully without any adverse event. The median operation time of Endoscopic Ruler was 3.00 min [interquartile range (IQR): 3.00 min]. The kappa value between Endoscopic Ruler and the endoscopists while detecting large OV was 0.52, demonstrating a moderate agreement. The kappa value for diagnosing large OV using Endoscopic Ruler among the six independent observers was 0.77, demonstrating a substantial agreement.

CONCLUSION

The data demonstrates that Endoscopic Ruler is feasible and safe for measuring the size of varices in patients with cirrhosis and portal hypertension. Endoscopic Ruler is potential to promote the clinical practice of the two-grade classification system of OV.

Estimation of Physiologic Pressures: Invasive and Non-Invasive Techniques, AI Models, and Future Perspectives

The measurement of physiologic pressure helps diagnose and prevent associated health complications. From typical conventional methods to more complicated modalities, such as the estimation of intracranial pressures, numerous invasive and noninvasive tools that provide us with insight into daily physiology and aid in understanding pathology are within our grasp. Currently, our standards for estimating vital pressures, including continuous BP measurements, pulmonary capillary wedge pressures, and hepatic portal gradients, involve the use of invasive modalities. As an emerging field in medical technology, artificial intelligence (AI) has been incorporated into analyzing and predicting patterns of physiologic pressures. AI has been used to construct models that have clinical applicability both in hospital settings and at-home settings for ease of use for patients. Studies applying AI to each of these compartmental pressures were searched and shortlisted for thorough assessment and review. There are several AI-based innovations in noninvasive blood pressure estimation based on imaging, auscultation, oscillometry and wearable technology employing biosignals. The purpose of this review is to provide an in-depth assessment of the involved physiologies, prevailing methodologies and emerging technologies incorporating AI in clinical practice for each type of compartmental pressure measurement. We also bring to the forefront AI-based noninvasive estimation techniques for physiologic pressure based on microwave systems that have promising potential for clinical practice.

Development and validation of a machine learning-based model for varices screening in compensated cirrhosis (CHESS2001): an international multicenter study

BACKGROUND AND AIMS

The prevalence of high-risk varices (HRV) is low among compensated cirrhotic patients undergoing EGD. Our study aimed to identify a novel machine learning (ML)-based model, named ML EGD, for ruling out HRV and avoiding unnecessary EGDs in patients with compensated cirrhosis.

METHODS

An international cohort from 17 institutions from China, Singapore, and India were enrolled (CHESS2001). The variables with the top 3 importance scores (liver stiffness, platelet count, and total bilirubin) were selected by the Shapley additive explanation and input into a light gradient-boosting machine algorithm to develop ML EGD for identification of HRV. Furthermore, we built a web-based calculator for ML EGD, which is free with open access (http://www.pan-chess.cn/calculator/MLEGD_score). Unnecessary EGDs that were not performed and the rates of missed HRV were used to assess the efficacy and safety for varices screening.

RESULTS

Of 2794 enrolled patients, 1283 patients formed a real-world cohort from 1 university hospital in China used to develop and internally validate the performance of ML EGD for varices screening. They were randomly assigned into the training (n = 1154) and validation (n = 129) cohorts with a ratio of 9:1. In the training cohort, ML EGD spared 607 (52.6%) unnecessary EGDs with a missed HRV rate of 3.6%. In the validation cohort, ML EGD spared 75 (58.1%) EGDs with a missed HRV rate of 1.4%. To externally test the performance of ML EGD, 966 patients from 14 university hospitals in China (test cohort 1) and 545 from 2 hospitals in Singapore and India (test cohort 2) comprised the 2 test cohorts. In test cohort 1, ML EGD spared 506 (52.4%) EGDs with a missed HRV rate of 2.8%. In test cohort 2, ML EGD spared 224 (41.1%) EGDs with a missed HRV rate of 3.1%. When compared with the Baveno VI criteria, ML EGD spared more screening EGDs in all cohorts (training cohort, 52.6% vs 29.4%; validation cohort, 58.1% vs 44.2%; test cohort 1, 52.4% vs 26.5%; test cohort 2, 41.1% vs 21.1%, respectively; P < .001).

CONCLUSIONS

We identified a novel model based on liver stiffness, platelet count, and total bilirubin, named ML EGD, as a free web-based calculator. ML EGD could efficiently help rule out HRV and avoid unnecessary EGDs in patients with compensated cirrhosis. (Clinical trial registration number: NCT04307264.).

A strategy for varices screening based on acoustic radiation force impulse combined with platelet (CHESS2001): An alternative of Baveno VI criteria

Few studies have reported on acoustic radiation force impulse (ARFI) for varices screening. Our study aimed to identify a strategy based on liver stiffness measurement (LSM) and spleen stiffness measurement (SSM) by ARFI combined with platelet count (PLT), named the ARP strategy, for ruling out high-risk varices (HRV) and avoiding unnecessary esophagogastroduodenoscopy (EGD) in patients with compensated cirrhosis. We retrospectively reviewed patients who underwent ARFI from a previous cohort (NCT04307264). Of them, patients between 2017 and 2019 composed the training cohort to develop the ARP strategy. The validation cohort consisted of others between 2015 and 2016 to validate and compare it with Baveno VI criteria about the performance for varices screening. Primary outcomes were the rates of spared EGDs and HRV missed. A total of 741 consecutive patients were included in the final analysis. Of them, 576 patients were included in the training cohort and 165 patients in the validation cohort. In the training cohort, ARP strategy was defined as LSM < 1.805 m/s or SSM < 2.445 m/s and PLT > 110 × 10⁹ /L. ARP strategy could spare 234 (40.6%) EGDs with a missed HRV rate of 3.4% (8 of 234). In the validation cohort, compared with Baveno VI criteria, the ARP strategy improved the proportion of avoided EGDs (49.7% vs. 34.5%; p < 0.001) and lowered the rate of misclassified HRV (1.2% vs. 3.5%; p < 0.001). Conclusion: The ARP strategy was an efficient and safe tool for varices screening in compensated cirrhosis, and it might be an auxiliary or even alternative to Baveno VI criteria.

Treatment Strategies in Emergency Endoscopy for Acute Esophageal Variceal Bleeding (CHESS1905): A Nationwide Cohort Study

Background and Aims

Emergency endoscopy is recommended for patients with acute esophageal variceal bleeding (EVB) and their prognosis has improved markedly over past decades due to the increased specialization of endoscopic practice. The study aimed to compare outcomes following emergency endoscopic injection sclerotherapy (EIS) and endoscopic variceal ligation (EVL) in cirrhotic patients with acute EVB.

Methods

Cirrhotic patients with acute EVB who underwent emergency endoscopy were retrospectively enrolled from 2013 to 2020 across 34 university hospitals from 30 cities. The primary outcome was the incidence of 5-day rebleeding after emergency endoscopy. Subgroup analysis was stratified by Child-Pugh class and bleeding history. A 1:1 propensity score matching (PSM) analysis was performed.

Results

A total of 1,017 and 382 patients were included in EIS group and EVL group, respectively. The 5-day rebleeding incidence was similar between EIS group and EVL group (4% vs. 5%, P = 0.45). The result remained the same after PSM (P = 1.00). Among Child-Pugh class A, B and C patients, there were no differences in the 5-day rebleeding incidence between the two groups after PSM (P = 0.25, 0.82, and 0.21, respectively). As for the patients with or without bleeding history, the differences between EIS group and EVL group were not significant after PSM (P = 1.00 and 0.26, respectively).

Conclusion

The nationwide cohort study indicates that EIS and EVL are both efficient emergency endoscopic treatment strategies for acute EVB. EIS should not be dismissed as an economical and effective emergency endoscopic treatment strategy of acute EVB. ClincialTrials.gov number NCT04307264.