Can we screen for pancreatic cancer? Identifying a sub-population of patients at high risk of subsequent diagnosis using machine learning techniques applied to primary care data

Machine Learning Models for Pancreatic Cancer Risk Prediction Using Electronic Health Record Data-A Systematic Review and Assessment

INTRODUCTION

Accurate risk prediction can facilitate screening and early detection of pancreatic cancer (PC). We conducted a systematic review to critically evaluate effectiveness of machine learning (ML) and artificial intelligence (AI) techniques applied to electronic health records (EHR) for PC risk prediction.

METHODS

Ovid MEDLINE(R), Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, Scopus, and Web of Science were searched for articles that utilized ML/AI techniques to predict PC, published between January 1, 2012, and February 1, 2024. Study selection and data extraction were conducted by 2 independent reviewers. Critical appraisal and data extraction were performed using the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies checklist. Risk of bias and applicability were examined using prediction model risk of bias assessment tool.

RESULTS

Thirty studies including 169,149 PC cases were identified. Logistic regression was the most frequent modeling method. Twenty studies utilized a curated set of known PC risk predictors or those identified by clinical experts. ML model discrimination performance (C-index) ranged from 0.57 to 1.0. Missing data were underreported, and most studies did not implement explainable-AI techniques or report exclusion time intervals.

DISCUSSION

AI/ML models for PC risk prediction using known risk factors perform reasonably well and may have near-term applications in identifying cohorts for targeted PC screening if validated in real-world data sets. The combined use of structured and unstructured EHR data using emerging AI models while incorporating explainable-AI techniques has the potential to identify novel PC risk factors, and this approach merits further study.

Risk factors for early-onset pancreatic ductal adenocarcinoma: A systematic literature review

BACKGROUND

Emerging cancer trends suggest an increase in pancreatic cancer incidence in individuals younger than its typical age of onset, potentially reflecting changes in population exposures and lifestyles.

PATIENTS AND METHODS

We conducted a PRISMA-standard systematic literature review to identify non-heritable risk factors for early-onset pancreatic ductal adenocarcinoma (PDAC) (PROSPERO number: CRD42022299397). Systematic searches of MEDLINE and Embase bibliographic databases were performed (January 2022), and publications were screened against predetermined eligibility criteria; data were extracted using standardised data fields. The STROBE checklist was used to assess the completeness of reporting as a proxy for publication quality. Data were categorised by risk factor and analysed descriptively.

RESULTS

In total, 24 publications were included. All publications reported observational study data; thresholds for age group comparisons ranged between 40 and 65 years. Lifestyle factors investigated included smoking, alcohol consumption, obesity, physical inactivity, meat intake, socioeconomic status and geographical residence. Clinical factors investigated included pancreatitis, diabetes/insulin resistance, prior cancer and cancer stage at diagnosis, hepatitis B infection, metabolic syndrome and long-term proton pump inhibitor exposure. Publication STROBE scores were 6-21 (maximum, 22). Eight studies reported results adjusted for confounders. Potential non-heritable risk factors for early-onset PDAC that warrant further investigation included smoking, alcohol consumption, pancreatitis and hepatitis B infection.

CONCLUSION

Evidence for non-heritable risk factors for early-onset PDAC is heterogeneous, but four factors were identified that might aid the identification of at-risk individuals who may benefit from screening and risk reduction strategies.

Role of artificial intelligence in the characterization of indeterminate pancreatic head mass and its usefulness in preoperative diagnosis

Artificial intelligence (AI) has been used in various fields of day-to-day life and its role in medicine is immense. Understanding of oncology has been improved with the introduction of AI which helps in diagnosis, treatment planning, management, prognosis, and follow-up. It also helps to identify high-risk groups who can be subjected to timely screening for early detection of malignant conditions. It is more important in pancreatic cancer as it is one of the major causes of cancer-related deaths worldwide and there are no specific early features (clinical and radiological) for diagnosis. With improvement in imaging modalities (computed tomography, magnetic resonance imaging, endoscopic ultrasound), most often clinicians were being challenged with lesions that were difficult to diagnose with human competence. AI has been used in various other branches of medicine to differentiate such indeterminate lesions including the thyroid gland, breast, lungs, liver, adrenal gland, kidney, etc. In the case of pancreatic cancer, the role of AI has been explored and is still ongoing. This review article will focus on how AI can be used to diagnose pancreatic cancer early or differentiate it from benign pancreatic lesions, therefore, management can be planned at an earlier stage.

An Integrative Pancreatic Cancer Risk Prediction Model in the UK Biobank

Pancreatic cancer (PaCa) is a lethal cancer with an increasing incidence, highlighting the need for early prevention strategies. There is a lack of a comprehensive PaCa predictive model derived from large prospective cohorts. Therefore, we have developed an integrated PaCa risk prediction model for PaCa using data from the UK Biobank, incorporating lifestyle-related, genetic-related, and medical history-related variables for application in healthcare settings. We used a machine learning-based random forest approach and a traditional multivariable logistic regression method to develop a PaCa predictive model for different purposes. Additionally, we employed dynamic nomograms to visualize the probability of PaCa risk in the prediction model. The top five influential features in the random forest model were age, PRS, pancreatitis, DM, and smoking. The significant risk variables in the logistic regression model included male gender (OR = 1.17), age (OR = 1.10), non-O blood type (OR = 1.29), higher polygenic score (PRS) (Q5 vs. Q1, OR = 2.03), smoking (OR = 1.82), alcohol consumption (OR = 1.27), pancreatitis (OR = 3.99), diabetes (DM) (OR = 2.57), and gallbladder-related disease (OR = 2.07). The area under the receiver operating curve (AUC) of the logistic regression model is 0.78. Internal validation and calibration performed well in both models. Our integrative PaCa risk prediction model with the PRS effectively stratifies individuals at future risk of PaCa, aiding targeted prevention efforts and supporting community-based cancer prevention initiatives.

The Utility of Artificial Intelligence in the Diagnosis and Management of Pancreatic Cancer

Artificial intelligence (AI) has made significant advancements in the medical domain in recent years. AI, an expansive field comprising Machine Learning (ML) and, within it, Deep Learning (DL), seeks to emulate the intricate operations of the human brain. It examines vast amounts of data and plays a crucial role in decision-making, overcoming limitations related to human evaluation. DL utilizes complex algorithms to analyze data. ML and DL are subsets of AI that utilize hard statistical techniques that help machines consistently improve at tasks with experience. Pancreatic cancer is more common in developed countries and is one of the leading causes of cancer-related mortality worldwide. Managing pancreatic cancer remains a challenge despite significant advancements in diagnosis and treatment. AI has secured an almost ubiquitous presence in the field of oncological workup and management, especially in gastroenterology malignancies. AI is particularly useful for various investigations of pancreatic carcinoma because it has specific radiological features that enable diagnostic procedures without the requirement of a histological study. However, interpreting and evaluating resulting images is not always simple since images vary as the disease progresses. Secondly, a number of factors may impact prognosis and response to the treatment process. Currently, AI models have been created for diagnosing, grading, staging, and predicting prognosis and treatment response. This review presents the most up-to-date knowledge on the use of AI in the diagnosis and treatment of pancreatic carcinoma.

A systematic review of clinical health conditions predicted by machine learning diagnostic and prognostic models trained or validated using real-world primary health care data

With the advances in technology and data science, machine learning (ML) is being rapidly adopted by the health care sector. However, there is a lack of literature addressing the health conditions targeted by the ML prediction models within primary health care (PHC) to date. To fill this gap in knowledge, we conducted a systematic review following the PRISMA guidelines to identify health conditions targeted by ML in PHC. We searched the Cochrane Library, Web of Science, PubMed, Elsevier, BioRxiv, Association of Computing Machinery (ACM), and IEEE Xplore databases for studies published from January 1990 to January 2022. We included primary studies addressing ML diagnostic or prognostic predictive models that were supplied completely or partially by real-world PHC data. Studies selection, data extraction, and risk of bias assessment using the prediction model study risk of bias assessment tool were performed by two investigators. Health conditions were categorized according to international classification of diseases (ICD-10). Extracted data were analyzed quantitatively. We identified 106 studies investigating 42 health conditions. These studies included 207 ML prediction models supplied by the PHC data of 24.2 million participants from 19 countries. We found that 92.4% of the studies were retrospective and 77.3% of the studies reported diagnostic predictive ML models. A majority (76.4%) of all the studies were for models' development without conducting external validation. Risk of bias assessment revealed that 90.8% of the studies were of high or unclear risk of bias. The most frequently reported health conditions were diabetes mellitus (19.8%) and Alzheimer's disease (11.3%). Our study provides a summary on the presently available ML prediction models within PHC. We draw the attention of digital health policy makers, ML models developer, and health care professionals for more future interdisciplinary research collaboration in this regard.

Early Diagnosis of Pancreatic Cancer via Machine Learning Analysis of a National Electronic Medical Record Database

PURPOSE

Pancreatic cancer (PaC) is often diagnosed at advanced stages, resulting in one of the lowest survival rates among patients with cancer. The purpose of this study was to investigate whether machine learning (ML) models can predict with high sensitivity and specificity an increased risk for PaC ahead of clinical diagnosis.

METHODS

Optum deidentified electronic health record (EHR) data set was used to extract 1-year data for each patient and to sample for PaC diagnosis, the number of interactions with the health care system, and unique demographic and clinical features. Data for patients with PaC diagnosis were collected between 1 and 2 years before the diagnosis. Standard binary classification ML models were used on training and testing data sets. Data analyses were performed using the scikit-learn package version 1.0.1.

RESULTS

The data set consisted of 18,987 patient EHRs collected between December 31, 2007, and December 31, 2017. EHRs with 10 unique features and at least three health care interactions were used for model training (N = 15,189; n = 8,438 [56%] with PaC) and testing (N = 3,798; n = 2,127 [56%] with PaC). The ensemble model achieved an AUC of 0.89, a sensitivity of 85.61%, and a specificity of 76.18% on the testing data set and produced superior results compared with other binary classifiers. Increasing unique health care interactions to nine failed to improve the AUC score. When the testing data set was enlarged to 5,696 patients, the ensemble model achieved an AUC of 0.92 and a specificity of 93.21%, but the sensitivity was compromised.

CONCLUSION

The ensemble model exceeded the state-of-the-art level of performance for prediction of PaC ahead of clinical diagnosis with a minimal clinically guided input, providing a potential strategy for selection of high-risk patients for further screening.

Real-Time Identification of Pancreatic Cancer Cases Using Artificial Intelligence Developed on Danish Nationwide Registry Data

PURPOSE

Pancreatic cancer is expected to be the second leading cause of cancer-related deaths worldwide within few years. Most patients are not diagnosed in time for curative-intent treatment. Accelerating the time of diagnosis is a key component of reducing pancreatic cancer mortality. We developed and tested a dynamic algorithm aiming at proactively identifying patients with a substantially elevated risk of having undiagnosed pancreatic cancer.

METHODS

Machine learning methodology was applied to a live stream of nationwide Danish registry data. A hybrid case-control and prospective cohort design relying on incidence density sampling was used. Three models with minimal tuning were tested. All performance evaluation metrics were based on out-of-sample, out-of-time data in a monthly walk-forward strategy to avoid any temporal biases or inflation of performance metrics. Outcome was a diagnosis of pancreatic cancer.

RESULTS

Subgroups identified had a 10.1% risk of being diagnosed with pancreatic cancer within 1 year, corresponding to a number needed to screen of 9.9. When considering competing, potentially computed tomography-detectable GI cancers, this number is reduced to 5.7. The time of diagnosis can be accelerated by up to 142 days.

CONCLUSION

Currently available nationwide live data and computational resources are sufficient for real-time identification of individuals with at least 10.1% risk of having undiagnosed pancreatic cancer and 17.7% risk of any GI cancer in the Danish population. For prospective identification of high-risk patients, the area under the curve is not a useful indication of the positive predictive values achieved. Viable design solutions are demonstrated, which address the main shortfalls of the existing cancer prediction efforts in relation to temporal biases, leaks, and performance metric inflation. Efficacy evaluations with resection rates and mortality as end points are needed.

A deep learning algorithm to predict risk of pancreatic cancer from disease trajectories

Pancreatic cancer is an aggressive disease that typically presents late with poor outcomes, indicating a pronounced need for early detection. In this study, we applied artificial intelligence methods to clinical data from 6 million patients (24,000 pancreatic cancer cases) in Denmark (Danish National Patient Registry (DNPR)) and from 3 million patients (3,900 cases) in the United States (US Veterans Affairs (US-VA)). We trained machine learning models on the sequence of disease codes in clinical histories and tested prediction of cancer occurrence within incremental time windows (CancerRiskNet). For cancer occurrence within 36 months, the performance of the best DNPR model has area under the receiver operating characteristic (AUROC) curve = 0.88 and decreases to AUROC (3m) = 0.83 when disease events within 3 months before cancer diagnosis are excluded from training, with an estimated relative risk of 59 for 1,000 highest-risk patients older than age 50 years. Cross-application of the Danish model to US-VA data had lower performance (AUROC = 0.71), and retraining was needed to improve performance (AUROC = 0.78, AUROC (3m) = 0.76). These results improve the ability to design realistic surveillance programs for patients at elevated risk, potentially benefiting lifespan and quality of life by early detection of this aggressive cancer.

Artificial Intelligence in the Diagnosis and Treatment of Pancreatic Cystic Lesions and Adenocarcinoma

Pancreatic cancer is projected to become the second leading cause of cancer-related mortality in the United States by 2030. This is in part due to the paucity of reliable screening and diagnostic options for early detection. Amongst known pre-malignant pancreatic lesions, pancreatic intraepithelial neoplasia (PanIN) and intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent. The current standard of care for the diagnosis and classification of pancreatic cystic lesions (PCLs) involves cross-sectional imaging studies and endoscopic ultrasound (EUS) and, when indicated, EUS-guided fine needle aspiration and cyst fluid analysis. However, this is suboptimal for the identification and risk stratification of PCLs, with accuracy of only 65-75% for detecting mucinous PCLs. Artificial intelligence (AI) is a promising tool that has been applied to improve accuracy in screening for solid tumors, including breast, lung, cervical, and colon cancer. More recently, it has shown promise in diagnosing pancreatic cancer by identifying high-risk populations, risk-stratifying premalignant lesions, and predicting the progression of IPMNs to adenocarcinoma. This review summarizes the available literature on artificial intelligence in the screening and prognostication of precancerous lesions in the pancreas, and streamlining the diagnosis of pancreatic cancer.

Automated classification of urine biomarkers to diagnose pancreatic cancer using 1-D convolutional neural networks

BACKGROUND

Early diagnosis of Pancreatic Ductal Adenocarcinoma (PDAC) is the main key to surviving cancer patients. Urine proteomic biomarkers which are creatinine, LYVE1, REG1B, and TFF1 present a promising non-invasive and inexpensive diagnostic method of the PDAC. Recent utilization of both microfluidics technology and artificial intelligence techniques enables accurate detection and analysis of these biomarkers. This paper proposes a new deep-learning model to identify urine biomarkers for the automated diagnosis of pancreatic cancers. The proposed model is composed of one-dimensional convolutional neural networks (1D-CNNs) and long short-term memory (LSTM). It can categorize patients into healthy pancreas, benign hepatobiliary disease, and PDAC cases automatically.

RESULTS

Experiments and evaluations have been successfully done on a public dataset of 590 urine samples of three classes, which are 183 healthy pancreas samples, 208 benign hepatobiliary disease samples, and 199 PDAC samples. The results demonstrated that our proposed 1-D CNN + LSTM model achieved the best accuracy score of 97% and the area under curve (AUC) of 98% versus the state-of-the-art models to diagnose pancreatic cancers using urine biomarkers.

CONCLUSION

A new efficient 1D CNN-LSTM model has been successfully developed for early PDAC diagnosis using four proteomic urine biomarkers of creatinine, LYVE1, REG1B, and TFF1. This developed model showed superior performance on other machine learning classifiers in previous studies. The main prospect of this study is the laboratory realization of our proposed deep classifier on urinary biomarker panels for assisting diagnostic procedures of pancreatic cancer patients.

Clinical Prediction Models for Pancreatic Cancer in General and At-Risk Populations: A Systematic Review

INTRODUCTION

Identifying high-risk individuals using a risk prediction model could be a crucial first stage of screening pathways to improve the early detection of pancreatic cancer. A systematic review was conducted to critically evaluate the published primary literature on the development or validation of clinical risk prediction models for pancreatic cancer risk.

METHODS

MEDLINE, Embase, and Web of Science were searched for relevant articles from the inception of each database up to November 2021. Study selection and data extraction were conducted by 2 independent reviewers. The Prediction model Risk Of Bias Assessment Tool (PROBAST) was applied to assess risk of bias.

RESULTS

In total, 33 studies were included, describing 38 risk prediction models. Excluding studies with an overlapping population, this study consist of 15,848,100 participants, of which 58,313 were diagnosed with pancreatic cancer. Eight studies externally validated their model, and 13 performed internal validation. The studies described risk prediction models for pancreatic cancer in the general population (n = 14), patients with diabetes (n = 8), and individuals with gastrointestinal (and other) symptoms (symptoms included abdominal pain, unexplained weight loss, jaundice, and change in bowel habits and indigestion; n = 11). The commonly used clinical risk factors in the model were cigarette smoking (n = 27), age (n = 25), diabetes history (n = 22), chronic pancreatitis (n = 18), and body mass index (n = 14). In the 25 studies that assessed model performance, C-statistics ranged from 0.61 to 0.98. Of the 33 studies included, 6 were rated as being at a low risk of bias based on PROBAST.

DISCUSSION

Many clinical risk prediction models for pancreatic cancer had been developed for different target populations. Although low risk-of-bias studies were identified, these require external validation and implementation studies to ensure that these will benefit clinical decision making.

Model-based screening for pancreatic cancer in Sweden

BACKGROUND

Detecting pancreatic cancer at an earlier stage may contribute to an increased survival. Patients with stage I pancreatic cancer have a 5-year survival rate of 36%, while stage IV patients have a 5-year survival rate of 1% in Sweden. Research into novel blood-based biomarkers for pancreatic cancer is highly intensive and innovative, but has yet to result in any routine screening test. The aim of this study was to evaluate the specificity and sensitivity of a hypothetical blood test for pancreatic cancer used for screening purposes and the economic aspects of testing.

METHOD

A model of a screening test was created, with varying specificity and sensitivity both set at 80%, 85%, 90%, 95% or 99% and applied to selected risk groups. Excessive costs of false positive screening outcomes, QALYs, ICERs and total costs were calculated.

RESULTS

Individuals with family history and genetic mutations associated with pancreatic cancer, new-onset diabetes ≥50 years of age and early symptoms had the highest positive predictive values and ICERs beneath the willingness-to-pay-level of EUR 100,000/QALY. Screening of the general population and smokers resulted in a high rate of false positive cases and extensive extra costs.

CONCLUSIONS

General screening for pancreatic cancer is not cost-effective, while screening of certain high-risk groups may be economically justified given the availability of a high-performing blood-based test.

BMI and HbA1c are metabolic markers for pancreatic cancer: Matched case-control study using a UK primary care database

Background

Weight loss, hyperglycaemia and diabetes are known features of pancreatic cancer. We quantified the timing and the amount of changes in body mass index (BMI) and glycated haemoglobin (HbA1c), and their association with pancreatic cancer from five years before diagnosis.

Methods

A matched case-control study was undertaken within 590 primary care practices in England, United Kingdom. 8,777 patients diagnosed with pancreatic cancer (cases) between 1^st January 2007 and 31^st August 2020 were matched to 34,979 controls by age, gender and diabetes. Longitudinal trends in BMI and HbA1c were visualised. Odds ratios adjusted for demographic and lifestyle factors (aOR) and 95% confidence intervals (CI) were calculated with conditional logistic regression. Subgroup analyses were undertaken according to the diabetes status.

Results

Changes in BMI and HbA1c observed for cases on longitudinal plots started one and two years (respectively) before diagnosis. In the year before diagnosis, a 1 kg/m² decrease in BMI between cases and controls was associated with aOR for pancreatic cancer of 1.05 (95% CI 1.05 to 1.06), and a 1 mmol/mol increase in HbA1c was associated with aOR of 1.06 (1.06 to 1.07). ORs remained statistically significant (p < 0.001) for 2 years before pancreatic cancer diagnosis for BMI and 3 years for HbA1c. Subgroup analysis revealed that the decrease in BMI was associated with a higher pancreatic cancer risk for people with diabetes than for people without (aORs 1.08, 1.06 to 1.09 versus 1.04, 1.03 to 1.05), but the increase in HbA1c was associated with a higher risk for people without diabetes than for people with diabetes (aORs 1.09, 1.07 to 1.11 versus 1.04, 1.03 to 1.04).

Conclusions

The statistically significant changes in weight and glycaemic control started three years before pancreatic cancer diagnosis but varied according to the diabetes status. The information from this study could be used to detect pancreatic cancer earlier than is currently achieved. However, regular BMI and HbA1c measurements are required to facilitate future research and implementation in clinical practice.

Artificial intelligence in pancreatic cancer

Pancreatic cancer is the deadliest disease, with a five-year overall survival rate of just 11%. The pancreatic cancer patients diagnosed with early screening have a median overall survival of nearly ten years, compared with 1.5 years for those not diagnosed with early screening. Therefore, early diagnosis and early treatment of pancreatic cancer are particularly critical. However, as a rare disease, the general screening cost of pancreatic cancer is high, the accuracy of existing tumor markers is not enough, and the efficacy of treatment methods is not exact. In terms of early diagnosis, artificial intelligence technology can quickly locate high-risk groups through medical images, pathological examination, biomarkers, and other aspects, then screening pancreatic cancer lesions early. At the same time, the artificial intelligence algorithm can also be used to predict the survival time, recurrence risk, metastasis, and therapy response which could affect the prognosis. In addition, artificial intelligence is widely used in pancreatic cancer health records, estimating medical imaging parameters, developing computer-aided diagnosis systems, etc. Advances in AI applications for pancreatic cancer will require a concerted effort among clinicians, basic scientists, statisticians, and engineers. Although it has some limitations, it will play an essential role in overcoming pancreatic cancer in the foreseeable future due to its mighty computing power.

A Comparison of Logistic Regression Against Machine Learning Algorithms for Gastric Cancer Risk Prediction Within Real-World Clinical Data Streams

PURPOSE

Noncardia gastric cancer (NCGC) is a leading cause of global cancer mortality, and is often diagnosed at advanced stages. Development of NCGC risk models within electronic health records (EHR) may allow for improved cancer prevention. There has been much recent interest in use of machine learning (ML) for cancer prediction, but few studies comparing ML with classical statistical models for NCGC risk prediction.

METHODS

We trained models using logistic regression (LR) and four commonly used ML algorithms to predict NCGC from age-/sex-matched controls in two EHR systems: Stanford University and the University of Washington (UW). The LR model contained well-established NCGC risk factors (intestinal metaplasia histology, prior Helicobacter pylori infection, race, ethnicity, nativity status, smoking history, anemia), whereas ML models agnostically selected variables from the EHR. Models were developed and internally validated in the Stanford data, and externally validated in the UW data. Hyperparameter tuning of models was achieved using cross-validation. Model performance was compared by accuracy, sensitivity, and specificity.

RESULTS

In internal validation, LR performed with comparable accuracy (0.732; 95% CI, 0.698 to 0.764), sensitivity (0.697; 95% CI, 0.647 to 0.744), and specificity (0.767; 95% CI, 0.720 to 0.809) to penalized lasso, support vector machine, K-nearest neighbor, and random forest models. In external validation, LR continued to demonstrate high accuracy, sensitivity, and specificity. Although K-nearest neighbor demonstrated higher accuracy and specificity, this was offset by significantly lower sensitivity. No ML model consistently outperformed LR across evaluation criteria.

CONCLUSION

Drawing data from two independent EHRs, we find LR on the basis of established risk factors demonstrated comparable performance to optimized ML algorithms. This study demonstrates that classical models built on robust, hand-chosen predictor variables may not be inferior to data-driven models for NCGC risk prediction.

Role of artificial intelligence in early detection and screening for pancreatic adenocarcinoma

Pancreatic adenocarcinoma remains to be one of the deadliest malignancies in the world despite treatment advancement over the past few decades. Its low survival rates and poor prognosis can be attributed to ambiguity in recommendations for screening and late symptom onset, contributing to its late presentation. In the recent years, artificial intelligence (AI) as emerged as a field to aid in the process of clinical decision making. Considerable efforts have been made in the realm of AI to screen for and predict future development of pancreatic ductal adenocarcinoma. This review discusses the use of AI in early detection and screening for pancreatic adenocarcinoma, and factors which may limit its use in a clinical setting.

Application of artificial intelligence to pancreatic adenocarcinoma

BACKGROUND AND OBJECTIVES

Pancreatic cancer (PC) is one of the deadliest cancers worldwide although substantial advancement has been made in its comprehensive treatment. The development of artificial intelligence (AI) technology has allowed its clinical applications to expand remarkably in recent years. Diverse methods and algorithms are employed by AI to extrapolate new data from clinical records to aid in the treatment of PC. In this review, we will summarize AI's use in several aspects of PC diagnosis and therapy, as well as its limits and potential future research avenues.

METHODS

We examine the most recent research on the use of AI in PC. The articles are categorized and examined according to the medical task of their algorithm. Two search engines, PubMed and Google Scholar, were used to screen the articles.

RESULTS

Overall, 66 papers published in 2001 and after were selected. Of the four medical tasks (risk assessment, diagnosis, treatment, and prognosis prediction), diagnosis was the most frequently researched, and retrospective single-center studies were the most prevalent. We found that the different medical tasks and algorithms included in the reviewed studies caused the performance of their models to vary greatly. Deep learning algorithms, on the other hand, produced excellent results in all of the subdivisions studied.

CONCLUSIONS

AI is a promising tool for helping PC patients and may contribute to improved patient outcomes. The integration of humans and AI in clinical medicine is still in its infancy and requires the in-depth cooperation of multidisciplinary personnel.

Developing practical clinical tools for predicting neonatal mortality at a neonatal intensive care unit in Tanzania

BACKGROUND

Neonatal mortality remains high in Tanzania at approximately 20 deaths per 1000 live births. Low birthweight, prematurity, and asphyxia are associated with neonatal mortality; however, no studies have assessed the value of combining underlying conditions and vital signs to provide clinicians with early warning of infants at risk of mortality. The aim of this study was to identify risk factors (including vital signs) associated with neonatal mortality in the neonatal intensive care unit (NICU) in Bugando Medical Centre (BMC), Mwanza, Tanzania; to identify the most accurate generalised linear model (GLM) or decision tree for predicting mortality; and to provide a tool that provides clinically relevant cut-offs for predicting mortality that is easily used by clinicians in a low-resource setting.

METHODS

In total, 165 neonates were enrolled between November 2019 and March 2020, of whom 80 (48.5%) died. We competed the performance of GLMs and decision trees by resampling the data to create training and test datasets and comparing their accuracy at correctly predicting mortality.

RESULTS

GLMs always outperformed decision trees. The best fitting GLM showed that (for standardised risk factors) temperature (OR 0.61, 95% CI 0.40-0.90), birthweight (OR 0.33, 95% CI 0.20-0.52), and oxygen saturation (OR 0.66, 95% CI 0.45-0.94) were negatively associated with mortality, while heart rate (OR 1.59, 95% CI 1.10-2.35) and asphyxia (OR 3.23, 95% 1.25-8.91) were risk factors. To identify the tool that balances accuracy and with ease of use in a low-resource clinical setting, we compared the best fitting GLM with simpler versions, and identified the three-variable GLM with temperature, heart rate, and birth weight as the best candidate. For this tool, cut-offs were identified using receiver operator characteristic (ROC) curves with the optimal cut-off for mortality prediction corresponding to 76.3% sensitivity and 68.2% specificity. The final tool is graphical, showing cut-offs that depend on birthweight, heart rate, and temperature.

CONCLUSIONS

Underlying conditions and vital signs can be combined into simple graphical tools that improve upon the current guidelines and are straightforward to use by clinicians in a low-resource setting.

Temporality of clinical factors associated with pancreatic cancer: a case-control study using linked electronic health records

BACKGROUND

Pancreatic cancer risk is poorly quantified in relation to the temporal presentation of medical comorbidities and lifestyle. This study aimed to examine this aspect, with possible influence of demographics.

METHODS

We conducted a retrospective case-control study on the ethnically-diverse population of East London, UK, using linked electronic health records. We evaluated the independent and two-way interaction effects of 19 clinico-demographic factors in patients with pancreatic cancer (N = 965), compared with non-malignant pancreatic conditions (N = 3963) or hernia (control; N = 4355), reported between April 1, 2008 and March 6, 2020. Risks were quantified by odds ratios (ORs) and 95% confidence intervals (CIs) from multivariable logistic regression models.

RESULTS

We observed increased odds of pancreatic cancer incidence associated with recent-onset diabetes occurring within 6 months to 3 years before cancer diagnosis (OR 1.95, 95% CI 1.25-3.03), long-standing diabetes for over 3 years (OR 1.74, 95% CI 1.32-2.29), recent smoking (OR 1.81, 95% CI 1.36-2.4) and drinking (OR 1.76, 95% CI 1.31-2.35), as compared to controls but not non-malignant pancreatic conditions. Pancreatic cancer odds was highest for chronic pancreatic disease patients (recent-onset: OR 4.76, 95% CI 2.19-10.3, long-standing: OR 5.1, 95% CI 2.18-11.9), amplified by comorbidities or harmful lifestyle. Concomitant diagnosis of diabetes, upper gastrointestinal or chronic pancreatic conditions followed by a pancreatic cancer diagnosis within 6 months were common, particularly in South Asians. Long-standing cardiovascular, respiratory and hepatobiliary conditions were associated with lower odds of pancreatic cancer.

CONCLUSIONS

Several factors are, independently or via effect modifications, associated with higher incidence of pancreatic cancer, but some established risk factors demonstrate similar magnitude of risk measures of developing non-malignant pancreatic conditions. The findings may inform refined risk-stratification strategies and better surveillance for high-risk individuals, and also provide a means for systematic identification of target population for prospective cohort-based early detection research initiatives.

Early Detection of Pancreatic Cancer: Applying Artificial Intelligence to Electronic Health Records

ABSTRACT

The potential of artificial intelligence (AI) applied to clinical data from electronic health records (EHRs) to improve early detection for pancreatic and other cancers remains underexplored. The Kenner Family Research Fund, in collaboration with the Cancer Biomarker Research Group at the National Cancer Institute, organized the workshop entitled: "Early Detection of Pancreatic Cancer: Opportunities and Challenges in Utilizing Electronic Health Records (EHR)" in March 2021. The workshop included a select group of panelists with expertise in pancreatic cancer, EHR data mining, and AI-based modeling. This review article reflects the findings from the workshop and assesses the feasibility of AI-based data extraction and modeling applied to EHRs. It highlights the increasing role of data sharing networks and common data models in improving the secondary use of EHR data. Current efforts using EHR data for AI-based modeling to enhance early detection of pancreatic cancer show promise. Specific challenges (biology, limited data, standards, compatibility, legal, quality, AI chasm, incentives) are identified, with mitigation strategies summarized and next steps identified.