Artificial neural network predicts the need for therapeutic ERCP in patients with suspected choledocholithiasis

Non-Invasive Prediction of Choledocholithiasis Using 1D Convolutional Neural Networks and Clinical Data

This paper introduces a novel one-dimensional convolutional neural network that utilizes clinical data to accurately detect choledocholithiasis, where gallstones obstruct the common bile duct. Swift and precise detection of this condition is critical to preventing severe complications, such as biliary colic, jaundice, and pancreatitis. This cutting-edge model was rigorously compared with other machine learning methods commonly used in similar problems, such as logistic regression, linear discriminant analysis, and a state-of-the-art random forest, using a dataset derived from endoscopic retrograde cholangiopancreatography scans performed at Olive View-University of California, Los Angeles Medical Center. The one-dimensional convolutional neural network model demonstrated exceptional performance, achieving 90.77% accuracy and 92.86% specificity, with an area under the curve of 0.9270. While the paper acknowledges potential areas for improvement, it emphasizes the effectiveness of the one-dimensional convolutional neural network architecture. The results suggest that this one-dimensional convolutional neural network approach could serve as a plausible alternative to endoscopic retrograde cholangiopancreatography, considering its disadvantages, such as the need for specialized equipment and skilled personnel and the risk of postoperative complications. The potential of the one-dimensional convolutional neural network model to significantly advance the clinical diagnosis of this gallstone-related condition is notable, offering a less invasive, potentially safer, and more accessible alternative.

Application of Deep Learning for Real-Time Ablation Zone Measurement in Ultrasound Imaging

BACKGROUND

The accurate delineation of ablation zones (AZs) is crucial for assessing radiofrequency ablation (RFA) therapy's efficacy. Manual measurement, the current standard, is subject to variability and potential inaccuracies.

AIM

This study aims to assess the effectiveness of Artificial Intelligence (AI) in automating AZ measurements in ultrasound images and compare its accuracy with manual measurements in ultrasound images.

METHODS

An in vitro study was conducted using chicken breast and liver samples subjected to bipolar RFA. Ultrasound images were captured every 15 s, with the AI model Mask2Former trained for AZ segmentation. The measurements were compared across all methods, focusing on short-axis (SA) metrics.

RESULTS

We performed 308 RFA procedures, generating 7275 ultrasound images across liver and chicken breast tissues. Manual and AI measurement comparisons for ablation zone diameters revealed no significant differences, with correlation coefficients exceeding 0.96 in both tissues (p < 0.001). Bland-Altman plots and a Deming regression analysis demonstrated a very close alignment between AI predictions and manual measurements, with the average difference between the two methods being -0.259 and -0.243 mm, for bovine liver and chicken breast tissue, respectively.

CONCLUSION

The study validates the Mask2Former model as a promising tool for automating AZ measurement in RFA research, offering a significant step towards reducing manual measurement variability.

Artificial intelligence: Applications in critical care gastroenterology

Gastrointestinal (GI) complications frequently necessitate intensive care unit (ICU) admission. Additionally, critically ill patients also develop GI complications requiring further diagnostic and therapeutic interventions. However, these patients form a vulnerable group, who are at risk for developing side effects and complications. Every effort must be made to reduce invasiveness and ensure safety of interventions in ICU patients. Artificial intelligence (AI) is a rapidly evolving technology with several potential applications in healthcare settings. ICUs produce a large amount of data, which may be employed for creation of AI algorithms, and provide a lucrative opportunity for application of AI. However, the current role of AI in these patients remains limited due to lack of large-scale trials comparing the efficacy of AI with the accepted standards of care.

A machine learning-based choledocholithiasis prediction tool to improve ERCP decision making: a proof-of-concept study

BACKGROUND

Previous studies demonstrated limited accuracy of existing guidelines for predicting choledocholithiasis, leading to overutilization of endoscopic retrograde cholangiopancreatography (ERCP). More accurate stratification may improve patient selection for ERCP and allow use of lower-risk modalities.

METHODS

A machine learning model was developed using patient information from two published cohort studies that evaluated performance of guidelines in predicting choledocholithiasis. Prediction models were developed using the gradient boosting model (GBM) machine learning method. GBM performance was evaluated using 10-fold cross-validation and area under the receiver operating characteristic curve (AUC). Important predictors of choledocholithiasis were identified based on relative importance in the GBM.

RESULTS

1378 patients (mean age 43.3 years; 61.2% female) were included in the GBM and 59.4% had choledocholithiasis. Eight variables were identified as predictors of choledocholithiasis. The GBM had accuracy of 71.5% (SD 2.5%) (AUC 0.79 [SD 0.06]) and performed better than the 2019 American Society for Gastrointestinal Endoscopy (ASGE) guidelines (accuracy 62.4% [SD 2.6%]; AUC 0.63 [SD 0.03]) and European Society of Gastrointestinal Endoscopy (ESGE) guidelines (accuracy 62.8% [SD 2.6%]; AUC 0.67 [SD 0.02]). The GBM correctly categorized 22% of patients directed to unnecessary ERCP by ASGE guidelines, and appropriately recommended as the next management step 48% of ERCPs incorrectly rejected by ESGE guidelines.

CONCLUSIONS

A machine learning-based tool was created, providing real-time, personalized, objective probability of choledocholithiasis and ERCP recommendations. This more accurately directed ERCP use than existing ASGE and ESGE guidelines, and has the potential to reduce morbidity associated with ERCP or missed choledocholithiasis.

A deep learning model for predicting multidrug-resistant organism infection in critically ill patients

BACKGROUND

This study aimed to apply the backpropagation neural network (BPNN) to develop a model for predicting multidrug-resistant organism (MDRO) infection in critically ill patients.

METHODS

This study collected patient information admitted to the intensive care unit (ICU) of the Affiliated Hospital of Qingdao University from August 2021 to January 2022. All patients enrolled were divided randomly into a training set (80%) and a test set (20%). The least absolute shrinkage and selection operator and stepwise regression analysis were used to determine the independent risk factors for MDRO infection. A BPNN model was constructed based on these factors. Then, we externally validated this model in patients from May 2022 to July 2022 over the same center. The model performance was evaluated by the calibration curve, the area under the curve (AUC), sensitivity, specificity, and accuracy.

RESULTS

In the primary cohort, 688 patients were enrolled, including 109 (15.84%) MDRO infection patients. Risk factors for MDRO infection, as determined by the primary cohort, included length of hospitalization, length of ICU stay, long-term bed rest, antibiotics use before ICU, acute physiology and chronic health evaluation II, invasive operation before ICU, quantity of antibiotics, chronic lung disease, and hypoproteinemia. There were 238 patients in the validation set, including 31 (13.03%) MDRO infection patients. This BPNN model yielded good calibration. The AUC of the training set, the test set and the validation set were 0.889 (95% CI 0.852-0.925), 0.919 (95% CI 0.856-0.983), and 0.811 (95% CI 0.731-0.891), respectively.

CONCLUSIONS

This study confirmed nine independent risk factors for MDRO infection. The BPNN model performed well and was potentially used to predict MDRO infection in ICU patients.

Acute Gallstone Pancreatitis: If a Picture Is Worth a Thousand Words, How Many Images Do We Need?

Introduction Accurate diagnosis and prompt definitive management of choledocholithiasis are vital in acute gallstone pancreatitis. The sensitivity of detection of choledocholithiasis varies across imaging modalities. Magnetic resonance cholangiopancreatography (MRCP) is the most sensitive but may not be necessary, resulting in both delayed definitive management and increased costs. We aimed to evaluate the range of radiological investigations patients with acute gallstone pancreatitis underwent and the clinical appropriateness of MRCP when performed. Methods This was an observational study of patients diagnosed with acute gallstone pancreatitis between January 1, 2019 and November 30, 2021 in a district general hospital in London, UK. A detailed review of patient records, laboratory and radiological results, and endoscopic and/or operative intervention was undertaken. Results One hundred consecutive patients diagnosed with acute gallstone pancreatitis (median age 57 years) were included. Seventy-nine had a transabdominal ultrasound (USS), 46 had CT, and 40 patients had MRCP. The median waiting time for these investigations was 1, 0, and 4 days, respectively. Choledocholithiasis was identified in 21 patients (4 on USS, 5 on CT, and 12 on MRCP). As definitive management, 37% underwent endoscopic retrograde cholangiopancreatography, and 57% underwent laparoscopic cholecystectomy. A total of 19% of patients were readmitted with pancreatitis prior to definitive management. Conclusions First-line imaging investigations such as USS and CT can detect some cases of choledocholithiasis in patients with acute gallstone pancreatitis, but not all. Despite expenses in terms of cost and length of hospital stay, MRCP remains an essential resource to detect cases of choledocholithiasis not captured by USS or CT. We recommend establishing a guideline to streamline imaging in assessing acute gallstone pancreatitis.

Test Performance Characteristics of Dynamic Liver Enzyme Trends in the Prediction of Choledocholithiasis

(1) Background: Various methods to predict the presence or absence of choledocholithiasis (CDL) have been proposed. We aimed to assess the performance characteristics of dynamic liver enzyme trends in the prediction of CDL. (2) Methods: This was a single-center retrospective cohort study. All adult in-patients undergoing endoscopy for suspected CDL between 1 January 2012 and 7 October 2018 were identified, with patients with prior cholecystectomy, prior sphincterotomy, or indwelling biliary prostheses were excluded. Available laboratory parameters within 72 h preceding the procedure were recorded, allowing for the assessment of trends. Dynamic enzyme trends were defined as any increase or decrease by 30% and 50% within 72 h of the index procedure. (3) Results: A total of 878 patients were included. Mean age was 61.8 years, with 58.6% female. Increases in alkaline phosphatase (ALP) of at least 30% or 50% were both specific for the presence of CDL, with specificities of 82.7% (95% CI 69.7–91.8%) and 88.5% (95% CI 76.6–95.6%), respectively. Decreases in bilirubin or ALP of at least 50% were highly specific for the absence of CDL, with specificities of 91.7% (95% CI 85.7–95.8%) and 100.0% (97.2–100.0%), respectively. (4) Conclusions: Several liver enzyme trends appear to be specific for the absence or presence of stones; in particular, significant decreases in total bilirubin or ALP of at least 30–50% over the prior 72 h appear to be especially predictive of an absence of intraductal findings during endoscopy.

An artificial intelligence algorithm is highly accurate for detecting endoscopic features of eosinophilic esophagitis

The endoscopic features associated with eosinophilic esophagitis (EoE) may be missed during routine endoscopy. We aimed to develop and evaluate an Artificial Intelligence (AI) algorithm for detecting and quantifying the endoscopic features of EoE in white light images, supplemented by the EoE Endoscopic Reference Score (EREFS). An AI algorithm (AI-EoE) was constructed and trained to differentiate between EoE and normal esophagus using endoscopic white light images extracted from the database of the University Hospital Augsburg. In addition to binary classification, a second algorithm was trained with specific auxiliary branches for each EREFS feature (AI-EoE-EREFS). The AI algorithms were evaluated on an external data set from the University of North Carolina, Chapel Hill (UNC), and compared with the performance of human endoscopists with varying levels of experience. The overall sensitivity, specificity, and accuracy of AI-EoE were 0.93 for all measures, while the AUC was 0.986. With additional auxiliary branches for the EREFS categories, the AI algorithm (AI-EoE-EREFS) performance improved to 0.96, 0.94, 0.95, and 0.992 for sensitivity, specificity, accuracy, and AUC, respectively. AI-EoE and AI-EoE-EREFS performed significantly better than endoscopy beginners and senior fellows on the same set of images. An AI algorithm can be trained to detect and quantify endoscopic features of EoE with excellent performance scores. The addition of the EREFS criteria improved the performance of the AI algorithm, which performed significantly better than endoscopists with a lower or medium experience level.

Artificial intelligence using advanced imaging techniques and cholangiocarcinoma: Recent advances and future direction

While cholangiocarcinoma represents only about 3% of all gastrointestinal tumors, it has a dismal survival rate, usually because it is diagnosed at a late stage. The utilization of Artificial Intelligence (AI) in medicine in general, and in gastroenterology has made gigantic steps. However, the application of AI for biliary disease, in particular for cholangiocarcinoma, has been sub-optimal. The use of AI in combination with clinical data, cross-sectional imaging (computed tomography, magnetic resonance imaging) and endoscopy (endoscopic ultrasound and cholangioscopy) has the potential to significantly improve early diagnosis and the choice of optimal therapeutic options, leading to a transformation in the prognosis of this feared disease. In this review we summarize the current knowledge on the use of AI for the diagnosis and management of cholangiocarcinoma and point to future directions in the field.

Artificial intelligence in the endoscopic approach of biliary tract diseases: A current review

In recent years there have been major developments in the field of artificial intelligence. The different areas of medicine have taken advantage of this tool to make various diagnostic and therapeutic methods more effective, safe, and user-friendly. In this way, artificial intelligence has been an increasingly present reality in medicine. In the field of Gastroenterology, the main application has been in the detection and characterization of colonic polyps, but an increasing number of studies have been published on the application of deep learning systems in other pathologies of the gastrointestinal tract. Evidence of the application of artificial intelligence in the assessment of biliary tract is still scarce. Some studies support the usefulness of these systems in the investigation and treatment of choledocholithiasis, demonstrating that they have the potential to be integrated into clinical practice and endoscopic procedures, such as endoscopic retrograde cholangiopancreatography. Its application in cholangioscopy for the investigation of undetermined biliary strictures also seems to be promising. Assessing the bile duct through endoscopic ultrasound can be challenging, especially for less experienced operators, thus becoming an area of potential interest for artificial intelligence. In this review, we summarize the state of the art of artificial intelligence in the endoscopic diagnosis and treatment of biliary diseases.

Endoscopic ultrasound avoids adverse events in high probability choledocholithiasis patients with a negative computed tomography

BACKGROUND

The current guideline recommends patients who meet high probability criteria for choledocholithiasis to receive endoscopic retrograde cholangiopancreatography (ERCP). However, adverse events can occur during ERCP. Our goal is to determine whether endoscopic ultrasound (EUS) before ERCP can avoid unnecessary ERCP complications, especially in patients with a negative CT scan.

METHODS

A total of 604 patients with high probability of choledocholithiasis were screened and 104 patients were prospectively enrolled. Patients with malignant biliary obstruction, altered GI anatomy, and choledocholithiasis on CT scan were excluded. Among them, 44 patients received EUS first, and ERCP if choledocholithiasis present (EUS-first group). The other 60 patients received ERCP directly (ERCP-first group). The baseline characteristics, presence of choledocholithiasis, and complications were compared between groups. All patients were followed for 3 months to determine the difference in recurrent biliary event rate. Cost-effectiveness was compared between the two strategies.

RESULTS

There was no marked difference in age, sex, laboratory data, presenting with pancreatitis, and risk factors for choledocholithiasis. Overall, 51 patients (49.0%) had choledocholithiasis, which did not justify the risk of direct ERCP. In the EUS-first group, 27 (61.4%) ERCP procedures were prevented. The overall complication rate was significantly lower in the EUS-first group compared to the ERCP-fist group (6.8% vs. 21.7%, P = 0.04). The number-needed-to-treat to avoid one unnecessary adverse event was 6.71. After a 3-month follow-up, the cumulative recurrence biliary event rates were similar (13.6% vs. 15.0%, P = 0.803). EUS-first strategy was more cost-effective than the ERCP-first strategy (mean cost 2322.89$ vs. 3175.63$, P = 0.002).

CONCLUSIONS

In high-probability choledocholithiasis patients with a negative CT, the EUS-first strategy is cost-effective, which can prevent unnecessary ERCP procedures and their complications.

Charting a path forward for clinical research in artificial intelligence and gastroenterology

Gastroenterology has been an early leader in bridging the gap between artificial intelligence (AI) model development and clinical trial validation, and in recent years we have seen the publication of several randomized clinical trials examining the role of AI in gastroenterology. As AI applications for clinical medicine advance rapidly, there is a clear need for guidance surrounding AI-specific study design, evaluation, comparison, analysis and reporting of results. Several initiatives are in the publication or pre-publication phase including AI-specific amendments to minimum reporting guidelines for clinical trials, society task force initiatives aimed at priority use cases and research priorities, and minimum reporting guidelines that guide the reporting of clinical prediction models. In this paper, we examine applications of AI in clinical trials and discuss elements of newly published AI-specific extensions to the Consolidated Standards of Reporting Trials and Standard Protocol Items: Recommendations for Interventional Trials statements that guide clinical trial reporting and development. We then review AI applications at the pre-trial level in both endoscopy and other subfields of gastroenterology and explore areas where further guidance is needed to supplement the current guidance available at the pre-trial level.

Machine learning models compared to existing criteria for noninvasive prediction of endoscopic retrograde cholangiopancreatography-confirmed choledocholithiasis

BACKGROUND AND AIMS

Noninvasive predictors of choledocholithiasis have generally exhibited marginal performance characteristics. We aimed to identify noninvasive independent predictors of endoscopic retrograde cholangiopancreatography (ERCP)-confirmed choledocholithiasis and accordingly developed predictive machine learning models (MLMs).

METHODS

Clinical data of consecutive patients undergoing first-ever ERCP for suspected choledocholithiasis from 2015-2019 were abstracted from a prospectively-maintained database. Multiple logistic regression was used to identify predictors of ERCP-confirmed choledocholithiasis. MLMs were then trained to predict ERCP-confirmed choledocholithiasis using pre-ERCP ultrasound (US) imaging only and separately using all available noninvasive imaging (US/CT/magnetic resonance cholangiopancreatography). The diagnostic performance of American Society for Gastrointestinal Endoscopy (ASGE) "high-likelihood" criteria was compared to MLMs.

RESULTS

We identified 270 patients (mean age 46 years, 62.2% female, 73.7% Hispanic/Latino, 59% with noninvasive imaging positive for choledocholithiasis) with native papilla who underwent ERCP for suspected choledocholithiasis, of whom 230 (85.2%) were found to have ERCP-confirmed choledocholithiasis. Logistic regression identified choledocholithiasis on noninvasive imaging (odds ratio (OR) = 3.045, P = 0.004) and common bile duct (CBD) diameter on noninvasive imaging (OR=1.157, P = 0.011) as predictors of ERCP-confirmed choledocholithiasis. Among the various MLMs trained, the random forest-based MLM performed best; sensitivity was 61.4% and 77.3% and specificity was 100% and 75.0%, using US-only and using all available imaging, respectively. ASGE high-likelihood criteria demonstrated sensitivity of 90.9% and specificity of 25.0%; using cut-points achieving this specificity, MLMs achieved sensitivity up to 97.7%.

CONCLUSIONS

MLMs using age, sex, race, presence of diabetes, fever, body mass index (BMI), total bilirubin, maximum CBD diameter, and choledocholithiasis on pre-ERCP noninvasive imaging predict ERCP-confirmed choledocholithiasis with good sensitivity and specificity and outperform the ASGE criteria for patients with suspected choledocholithiasis.

Artificial intelligence in gastroenterology: A state-of-the-art review

The development of artificial intelligence (AI) has increased dramatically in the last 20 years, with clinical applications progressively being explored for most of the medical specialties. The field of gastroenterology and hepatology, substantially reliant on vast amounts of imaging studies, is not an exception. The clinical applications of AI systems in this field include the identification of premalignant or malignant lesions (e.g., identification of dysplasia or esophageal adenocarcinoma in Barrett’s esophagus, pancreatic malignancies), detection of lesions (e.g., polyp identification and classification, small-bowel bleeding lesion on capsule endoscopy, pancreatic cystic lesions), development of objective scoring systems for risk stratification, predicting disease prognosis or treatment response [e.g., determining survival in patients post-resection of hepatocellular carcinoma), determining which patients with inflammatory bowel disease (IBD) will benefit from biologic therapy], or evaluation of metrics such as bowel preparation score or quality of endoscopic examination. The objective of this comprehensive review is to analyze the available AI-related studies pertaining to the entirety of the gastrointestinal tract, including the upper, middle and lower tracts; IBD; the hepatobiliary system; and the pancreas, discussing the findings and clinical applications, as well as outlining the current limitations and future directions in this field.

Systematic literature review of machine learning methods used in the analysis of real-world data for patient-provider decision making

BACKGROUND

Machine learning is a broad term encompassing a number of methods that allow the investigator to learn from the data. These methods may permit large real-world databases to be more rapidly translated to applications to inform patient-provider decision making.

METHODS

This systematic literature review was conducted to identify published observational research of employed machine learning to inform decision making at the patient-provider level. The search strategy was implemented and studies meeting eligibility criteria were evaluated by two independent reviewers. Relevant data related to study design, statistical methods and strengths and limitations were identified; study quality was assessed using a modified version of the Luo checklist.

RESULTS

A total of 34 publications from January 2014 to September 2020 were identified and evaluated for this review. There were diverse methods, statistical packages and approaches used across identified studies. The most common methods included decision tree and random forest approaches. Most studies applied internal validation but only two conducted external validation. Most studies utilized one algorithm, and only eight studies applied multiple machine learning algorithms to the data. Seven items on the Luo checklist failed to be met by more than 50% of published studies.

CONCLUSIONS

A wide variety of approaches, algorithms, statistical software, and validation strategies were employed in the application of machine learning methods to inform patient-provider decision making. There is a need to ensure that multiple machine learning approaches are used, the model selection strategy is clearly defined, and both internal and external validation are necessary to be sure that decisions for patient care are being made with the highest quality evidence. Future work should routinely employ ensemble methods incorporating multiple machine learning algorithms.

Barrett esophagus: What to expect from Artificial Intelligence?

The evaluation and assessment of Barrett's esophagus is challenging for both expert and nonexpert endoscopists. However, the early diagnosis of cancer in Barrett's esophagus is crucial for its prognosis, and could save costs. Pre-clinical and clinical studies on the application of Artificial Intelligence (AI) in Barrett's esophagus have shown promising results. In this review, we focus on the current challenges and future perspectives of implementing AI systems in the management of patients with Barrett's esophagus.

Applications of artificial intelligence in pancreatic and biliary diseases

The application of artificial intelligence (AI) in medicine has increased rapidly with respect to tasks including disease detection/diagnosis, risk stratification, and prognosis prediction. With recent advances in computing power and algorithms, AI has shown promise in taking advantage of vast electronic health data and imaging studies to supplement clinicians. Machine learning and deep learning are the most widely used AI methodologies for medical research and have been applied in pancreatobiliary diseases for which diagnosis and treatment selection are often complicated and require joint consideration of data from multiple sources. The aim of this review is to provide a concise introduction of the major AI methodologies and the current landscape of AI research in pancreatobiliary diseases.

Artificial intelligence in pancreaticobiliary endoscopy

Artificial intelligence (AI) applications in health care have exponentially increased in recent years, and a few of these are related to pancreatobiliary disorders. AI-based methods were applied to extract information, in prognostication, to guide clinical treatment decisions and in pancreatobiliary endoscopy to characterize lesions. AI applications in endoscopy are expected to reduce inter-operator variability, improve the accuracy of diagnosis, and assist in therapeutic decision-making in real time. AI-based literature must however be interpreted with caution given the limited external validation. A multidisciplinary approach combining clinical and imaging or endoscopy data will better utilize AI-based technologies to further improve patient care.

Application of artificial intelligence in pancreaticobiliary diseases

The role of artificial intelligence and its applications has been increasing at a rapid pace in the field of gastroenterology. The application of artificial intelligence in gastroenterology ranges from colon cancer screening and characterization of dysplastic and neoplastic polyps to the endoscopic ultrasonographic evaluation of pancreatic diseases. Artificial intelligence has been found to be useful in the evaluation and enhancement of the quality measure for endoscopic retrograde cholangiopancreatography. Similarly, artificial intelligence techniques like artificial neural networks and faster region-based convolution network are showing promising results in early and accurate diagnosis of pancreatic cancer and its differentiation from chronic pancreatitis. Other artificial intelligence techniques like radiomics-based computer-aided diagnosis systems could help to differentiate between various types of cystic pancreatic lesions. Artificial intelligence and computer-aided systems also showing promising results in the diagnosis of cholangiocarcinoma and the prediction of choledocholithiasis. In this review, we discuss the role of artificial intelligence in establishing diagnosis, prognosis, predicting response to treatment, and guiding therapeutics in the pancreaticobiliary system.

Artificial intelligence in biliopancreatic endoscopy: Is there any role?

Artificial intelligence (AI) research in endoscopy is being translated at rapid pace with a number of approved devices now available for use in luminal endoscopy. However, the published literature for AI in biliopancreatic endoscopy is predominantly limited to early pre-clinical studies including applications for diagnostic EUS and patient risk stratification. Potential future use cases are highlighted in this manuscript including optical characterisation of strictures during cholangioscopy, prediction of post-ERCP acute pancreatitis and selective biliary duct cannulation difficulty, automated report generation and novel AI-based quality key performance metrics. To realise the full potential of AI and accelerate innovation, it is crucial that robust inter-disciplinary collaborations are formed between biliopancreatic endoscopists and AI researchers.

The Challenges of Implementing Artificial Intelligence into Surgical Practice

BACKGROUND

Artificial intelligence is touted as the future of medicine. Classical algorithms for the detection of common bile duct stones (CBD) have had poor clinical uptake due to low accuracy. This study explores the challenges of developing and implementing a machine-learning model for the prediction of CBD stones in patients presenting with acute biliary disease (ABD).

METHODS

All patients presenting acutely to Christchurch Hospital over a two-year period with ABD were retrospectively identified. Clinical data points including lab test results, demographics and ethnicity were recorded. Several statistical techniques were utilised to develop a machine-learning model. Issues with data collection, quality, interpretation and barriers to implementation were identified and highlighted.

RESULTS

Issues with patient identification, coding accuracy, and implementation were encountered. In total, 1315 patients met inclusion criteria. Incorrect international classification of disease 10 (ICD-10) coding was noted in 36% (137/382) of patients recorded as having CBD stones. Patients with CBD stones were significantly older and had higher aspartate aminotransferase (AST), alanine aminotransferase (ALT), bilirubin and gamma-glutamyl transferase (GGT) levels (p < 0.001). The no information rate was 81% (1070/1315 patients). The optimum model developed was the gradient boosted model with a PPV of 67%, NPV of 87%, sensitivity of 37% and a specificity of 96% for common bile duct stones.

CONCLUSION

This paper highlights the utility of machine learning in predicting CBD stones. Accuracy is limited by current data and issues do exist around both the ethics and practicality of implementation. Regardless, machine learning represents a promising new paradigm for surgical practice.

Application of Artificial Intelligence to Gastroenterology and Hepatology

Since 2010, substantial progress has been made in artificial intelligence (AI) and its application to medicine. AI is explored in gastroenterology for endoscopic analysis of lesions, in detection of cancer, and to facilitate the analysis of inflammatory lesions or gastrointestinal bleeding during wireless capsule endoscopy. AI is also tested to assess liver fibrosis and to differentiate patients with pancreatic cancer from those with pancreatitis. AI might also be used to establish prognoses of patients or predict their response to treatments, based on multiple factors. We review the ways in which AI may help physicians make a diagnosis or establish a prognosis and discuss its limitations, knowing that further randomized controlled studies will be required before the approval of AI techniques by the health authorities.

ASGE guideline on the role of endoscopy in the evaluation and management of choledocholithiasis

Each year choledocholithiasis results in biliary obstruction, cholangitis, and pancreatitis in a significant number of patients. The primary treatment, ERCP, is minimally invasive but associated with adverse events in 6% to 15%. This American Society for Gastrointestinal Endoscopy (ASGE) Standard of Practice (SOP) Guideline provides evidence-based recommendations for the endoscopic evaluation and treatment of choledocholithiasis. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework was used to rigorously review and synthesize the contemporary literature regarding the following topics: EUS versus MRCP for diagnosis, the role of early ERCP in gallstone pancreatitis, endoscopic papillary dilation after sphincterotomy versus sphincterotomy alone for large bile duct stones, and impact of ERCP-guided intraductal therapy for large and difficult choledocholithiasis. Comprehensive systematic reviews were also performed to assess the following: same-admission cholecystectomy for gallstone pancreatitis, clinical predictors of choledocholithiasis, optimal timing of ERCP vis-à-vis cholecystectomy, management of Mirizzi syndrome and hepatolithiasis, and biliary stent therapy for choledocholithiasis. Core clinical questions were derived using an iterative process by the ASGE SOP Committee. This body developed all recommendations founded on the certainty of the evidence, balance of risks and harms, consideration of stakeholder preferences, resource utilization, and cost-effectiveness.

Artificial Neural Networking Model for the Prediction of Early Occlusion of Bilateral Plastic Stent Placement for Inoperable Hilar Cholangiocarcinoma

BACKGROUND

This study aimed to determine whether the back-propagation artificial neural network (BP-ANN) model could be constructed to accurately in predicting early occlusion of bilateral plastic stent placement for inoperable hilar cholangiocarcinoma (HCA).

METHODS

A total of 288 patients from the An Hui provincial Hospital were randomly divided into the training cohort (80%) and the internal testing cohort (20%). The predictive accuracy of the BP-ANN for predicting early occlusion of bilateral plastic stent placement of inoperable HCA was measured by the area under the curve (AUC) on receiver operating characteristic (ROC) curve analysis. The results were compared with those obtained using the conventional multivariate logistic regression analysis.

RESULTS

Multivariate analysis revealed that cancer stage (P=0.005) and Bismuth stage (P=0.003) were independently and significantly associated with early stent occlusion. In the training cohort, BP-ANN had larger AUC than the multivariate logistic regression model (P=0.00049). In the internal testing cohort, the AUC of the BP-ANN had larger AUC than the multivariate logistic regression model (P=0.02142).

CONCLUSIONS

This study showed that the BP-ANN model is a good predictive tool. It performed better than the conventional and commonly used statistical model in predicting early occlusion of bilateral plastic stent placement for inoperable HCA.

Modeling of glucose release from native and modified wheat starch gels during in vitro gastrointestinal digestion using artificial intelligence methods

Estimation of the amounts of glucose release (AGR) during gastrointestinal digestion can be useful to identify food of potential use in the diet of individuals with diabetes. In this work, adaptive neuro-fuzzy inference system (ANFIS), genetic algorithm-artificial neural network (GA-ANN) and group method of data handling (GMDH) models were applied to estimate the AGR from native (NWS), cross-linked (CLWS) and hydroxypropylated wheat starch (HPWS) gels during digestion under simulated gastrointestinal conditions. The GA-ANN and ANFIS were fed with 3 inputs of digestion time (1-120min), gel volume (7.5 and 15ml) and concentration (8 and 12%, w/w) for prediction of the AGR. The developed ANFIS predictions were close to the experimental data (r=0.977-0.996 and RMSE=0.225-0.619). The optimized GA-ANN, which included 6-7 hidden neurons, predicted the AGR with a good precision (r=0.984-0.993 and RMSE=0.338-0.588). Also, a three layers GMDH model with 3 neurons accurately predicted the AGR (r=0.979-0.986 and RMSE=0.339-0.443). Sensitivity analysis data demonstrated that the gel concentration was the most sensitive factor for prediction of the AGR. The results dedicated that the AGR will be accurately predictable through such soft computing methods providing less computational cost and time.

Predicting the likelihood of a persistent bile duct stone in patients with suspected choledocholithiasis: accuracy of existing guidelines and the impact of laboratory trends

BACKGROUND

Existing guidelines aim to stratify the likelihood of choledocholithiasis to guide the use of ERCP versus a lower-risk diagnostic study such as EUS, MRCP, or intraoperative cholangiography.

OBJECTIVE

To assess the performance of existing guidelines in predicting choledocholithiasis and to determine whether trends in laboratory parameters improve diagnostic accuracy.

DESIGN

Retrospective cohort study.

SETTING

Tertiary-care hospital.

PATIENTS

Hospitalized patients presenting with suspected choledocholithiasis over a 6-year period.

INTERVENTIONS

Assessment of the American Society for Gastrointestinal Endoscopy (ASGE) guidelines, its component variables, and laboratory trends in predicting choledocholithiasis.

MAIN OUTCOME MEASUREMENTS

The presence of choledocholithiasis confirmed by EUS, MRCP, or ERCP.

RESULTS

A total of 179 (35.9%) of the 498 eligible patients met ASGE high-probability criteria for choledocholithiasis on initial presentation. Of those, 99 patients (56.3%) had a stone/sludge on subsequent confirmatory test. Of patients not meeting high-probability criteria on presentation, 111 (34.8%) had a stone/sludge. The overall accuracy of the guidelines in detecting choledocholithiasis was 62.1% (47.4% sensitivity, 73% specificity) based on data available at presentation. The accuracy was unchanged when incorporating the second set of liver chemistries obtained after admission (63.2%), suggesting that laboratory trends do not improve performance.

LIMITATIONS

Retrospective study, inconsistent timing of the second set of biochemical markers.

CONCLUSION

In our cohort of patients, existing choledocholithiasis guidelines lacked diagnostic accuracy, likely resulting in overuse of ERCP. Incorporation of laboratory trends did not improve performance. Additional research focused on risk stratification is necessary to meet the goal of eliminating unnecessary diagnostic ERCP.

Regression tree for choledocholithiasis prediction

OBJECTIVES

The aim of this study was to develop and compare the predictive accuracy of classification and regression tree (CART) analysis with logistic regression (LR) for predicting common bile duct stones (CBDS) in patients subjected to laparoscopic cholecystectomy.

PATIENTS AND METHODS

We prospectively collected preoperative (demographic, biochemical, ultrasonographic) and intraoperative (intraoperative cholangiography, cystic duct diameter) data for 154 patients considered for elective laparoscopic cholecystectomy at the department of General Surgery at Gornji Milanovac from 2013 through 2014. Univariate and multivariate regression analyses were used to determine independent predictors of CBDS. The CART analysis was carried out using the predictors identified by LR analysis. Various measures for the assessment of risk prediction models were determined, such as predictive ability, accuracy, the area under the receiver operating characteristic curve, and clinical utility using decision curve analysis.

RESULTS

The most decisive variable at the time of classification was the cystic duct diameter category, the alkaline phosphatase, and dangerous stones. The CART model was shown to have good discriminatory ability (93.9%). Accuracy was similar in both models, ranging from 92.9% in the CART model and 93.5% in the LR model. In decision curve analysis, the CART model outperformed the LR model.

CONCLUSION

We developed a user-friendly risk model that can successfully predict the presence of choledocholithiasis in patients planned for elective cholecystectomy. However, before recommending its use in clinical practice, a larger and more complete database should be used to further clarify the differences between models in terms of prediction of the CBDS.