GCdiscrimination: identification of gastric cancer based on a milliliter of blood

Non-endoscopic Applications of Machine Learning in Gastric Cancer: A Systematic Review

PURPOSE

Gastric cancer is an important health burden characterized by high prevalence and mortality rate. Upper gastrointestinal endoscopy coupled with biopsy is the primary means in which gastric cancer is diagnosed, and most of machine learning (ML) tools are developed in this area. This systematic review focuses on the applications of ML in gastric cancer that do not involve endoscopic image recognition.

METHODS

A systematic review of ML applications that do not involve endoscopy and are relevant to gastric cancer was performed in two databases and independently evaluated by the two authors. Information collected from the included studies are year of publication, ML algorithm, ML performance, specimen used to create the ML model, and clinical application of the model.

RESULTS

From 791 screened studies, 63 studies were included in the systematic review. The included studies demonstrate that the non-endoscopic applications of ML can be divided into three main categories, which are diagnostics, predicting response to therapy, and prognosis prediction. Various specimen and algorithms were found to be used for these applications. Most of its clinical use includes histopathologic slide reading in the diagnosis of gastric cancer and a risk scoring system to determine the survival of patients or to determine the important variables that may affect the patient's prognosis.

CONCLUSION

The systematic review suggests that there are numerous examples of non-endoscopic applications of ML that are relevant to gastric cancer. These studies have utilized various specimens, even non-conventional ones, thus showing great promise for the development of more non-invasive techniques. However, most of these studies are still in the early stages and will take more time before they can be clinically deployed. Moving forward, researchers in this field of study are encouraged to improve data curation and annotation, improve model interpretability, and compare model performance with the currently accepted standard in the clinical practice.

Establishment of a differential diagnosis method and an online prediction platform for AOSD and sepsis based on gradient boosting decision trees algorithm

OBJECTIVE

The differential diagnosis between adult-onset Still's disease (AOSD) and sepsis has always been a challenge. In this study, a machine learning model for differential diagnosis of AOSD and sepsis was developed and an online platform was developed to facilitate the clinical application of the model.

METHODS

All data were collected from 42 AOSD patients and 50 sepsis patients admitted to Affiliated Hospital of Xuzhou Medical University from December 2018 to December 2021. In addition, 5 AOSD patients and 10 sepsis patients diagnosed in our hospital after March 2022 were collected for external validation. All models were built using the scikit-learn library (version 1.0.2) in Python (version 3.9.7), and feature selection was performed using the SHAP (Shapley Additive exPlanation) package developed in Python.

RESULTS

The results showed that the gradient boosting decision tree(GBDT) optimization model based on arthralgia, ferritin × lymphocyte count, white blood cell count, ferritin × platelet count, and α1-acid glycoprotein/creatine kinase could well identify AOSD and sepsis. The training set interaction test (AUC: 0.9916, ACC: 0.9457, Sens: 0.9556, Spec: 0.9578) and the external validation also achieved satisfactory results (AUC: 0.9800, ACC: 0.9333, Sens: 0.8000, Spec: 1.000). We named this discrimination method AIADSS (AI-assisted discrimination of Still's disease and Sepsis) and created an online service platform for practical operation, the website is http://cppdd.cn/STILL1/ .

CONCLUSION

We created a method for the identification of AOSD and sepsis based on machine learning. This method can provide a reference for clinicians to formulate the next diagnosis and treatment plan.

A novel nomogram based on the hematological prognosis risk scoring system can predict the overall survival of patients with hepatocellular carcinoma

BACKGROUND

This study aimed to establish and validate a nomogram based on a hematological prognostic risk scoring system to predict the overall survival in patients with unresectable hepatocellular carcinoma (HCC).

MATERIALS AND METHODS

Patients diagnosed with unresectable HCC undergoing transcatheter arterial chemoembolization (TACE) in 2012-2016 and 2017-2018 were included in the development set and validation set, respectively. The clinical outcome was overall survival (OS). The LASSO regression analysis was used to construct a hematological prognostic risk scoring system (HPR) by using the 18 hematological markers of patients in the development set. Combining the features of oncology on the basis of HPR to construct a nomogram for OS. In the development set and validation sets, the C-index, calibration curve, and decision curve analysis (DCA) were used to evaluate the prediction performance of the nomogram.

RESULTS

Multiple markers of immunity, coagulation, liver function, and nutrition, including red blood cell distribution width-coefficient of variation (RDW-CV), platelet (PLT), aspartate transferase (AST), alkaline phosphatase (ALP), prognostic nutritional index (PNI), and fibrinogen (Fib), construct the HPR. HPR was an independent risk factor for OS in patients with HCC. The C-index of the nomogram was 0.731 (95% confidence interval (CI) 0.712-0.749) and 0.696 (95% CI 0.668-0.725) in the development set and the validation set, respectively.

CONCLUSIONS

HPR was a complement to the clinical features of patients with unresectable HCC. The nomogram based on HPR proved to be a practical and effective method for prognosticating HCC patients who undergo TACE.

Sequence-based machine learning method for predicting the effects of phosphorylation on protein-protein interactions

Protein phosphorylation, catalyzed by kinases, is an important biochemical process, which plays an essential role in multiple cell signaling pathways. Meanwhile, protein-protein interactions (PPI) constitute the signaling pathways. Abnormal phosphorylation status on protein can regulate protein functions through PPI to evoke severe diseases, such as Cancer and Alzheimer's disease. Due to the limited experimental evidence and high costs to experimentally identify novel evidence of phosphorylation regulation on PPI, it is necessary to develop a high-accuracy and user-friendly artificial intelligence method to predict phosphorylation effect on PPI. Here, we proposed a novel sequence-based machine learning method named PhosPPI, which achieved better identification performance (Accuracy and AUC) than other competing predictive methods of Betts, HawkDock and FoldX. PhosPPI is now freely available in web server (https://phosppi.sjtu.edu.cn/). This tool can help the user to identify functional phosphorylation sites affecting PPI and explore phosphorylation-associated disease mechanism and drug development.

A computational framework of routine test data for the cost-effective chronic disease prediction

Chronic diseases, because of insidious onset and long latent period, have become the major global disease burden. However, the current chronic disease diagnosis methods based on genetic markers or imaging analysis are challenging to promote completely due to high costs and cannot reach universality and popularization. This study analyzed massive data from routine blood and biochemical test of 32 448 patients and developed a novel framework for cost-effective chronic disease prediction with high accuracy (AUC 87.32%). Based on the best-performing XGBoost algorithm, 20 classification models were further constructed for 17 types of chronic diseases, including 9 types of cancers, 5 types of cardiovascular diseases and 3 types of mental illness. The highest accuracy of the model was 90.13% for cardia cancer, and the lowest was 76.38% for rectal cancer. The model interpretation with the SHAP algorithm showed that CREA, R-CV, GLU and NEUT% might be important indices to identify the most chronic diseases. PDW and R-CV are also discovered to be crucial indices in classifying the three types of chronic diseases (cardiovascular disease, cancer and mental illness). In addition, R-CV has a higher specificity for cancer, ALP for cardiovascular disease and GLU for mental illness. The association between chronic diseases was further revealed. At last, we build a user-friendly explainable machine-learning-based clinical decision support system (DisPioneer: http://bioinfor.imu.edu.cn/dispioneer) to assist in predicting, classifying and treating chronic diseases. This cost-effective work with simple blood tests will benefit more people and motivate clinical implementation and further investigation of chronic diseases prevention and surveillance program.

Prognostic significance of pretreatment red blood cell distribution width in primary diffuse large B-cell lymphoma of the central nervous system for 3P medical approaches in multiple cohorts

Background/aims

Predicting the clinical outcomes of primary diffuse large B-cell lymphoma of the central nervous system (PCNS-DLBCL) to methotrexate-based combination immunochemotherapy treatment in advance and therefore administering the tailored treatment to the individual is consistent with the principle of predictive, preventive, and personalized medicine (PPPM/3PM). The red blood cell distribution width (RDW) has been reported to be associated with the clinical outcomes of multiple cancer. However, its prognostic role in PCNS-DLBCL is yet to be evaluated. Therefore, we aimed to effectively stratify PCNS-DLBCL patients with different prognosis in advance and early identify the patients who were appropriate to methotrexate-based combination immunochemotherapy based on the pretreatment level of RDW and a clinical prognostic model.

Methods

A prospective-retrospective, multi-cohort study was conducted from 2010 to 2020. We evaluated RDW in 179 patients (retrospective discovery cohorts of Huashan Center and Renji Center and prospective validation cohort of Cancer Center) with PCNS-DLBCL treated with methotrexate-based combination immunochemotherapy. A generalized additive model with locally estimated scatterplot smoothing was used to identify the relationship between pretreatment RDW levels and clinical outcomes. The high vs low risk of RDW combined with MSKCC score was determined by a minimal P-value approach. The clinical outcomes in different groups were then investigated.

Results

The pretreatment RDW showed a U-shaped relationship with the risk of overall survival (OS, P = 0.047). The low RDW (< 12.6) and high RDW (> 13.4) groups showed significantly worse OS (P < 0.05) and progression-free survival (PFS; P < 0.05) than the median group (13.4 > RDW > 12.6) in the discovery and validation cohort, respectively. RDW could predict the clinical outcomes successfully. In the discovery cohort, RDW achieved the area under the receiver operating characteristic curve (AUC) of 0.9206 in predicting the clinical outcomes, and the predictive value (AUC = 0.7177) of RDW was verified in the validation cohort. In addition, RDW combined with MSKCC predictive model can distinguish clinical outcomes with the AUC of 0.8348 for OS and 0.8125 for PFS. Compared with the RDW and MSKCC prognosis variables, the RDW combined with MSKCC scores better identified a subgroup of patients with favorable long-term survival in the validation cohort (P < 0.001). RDW combined MSKCC score remained to be independently associated with clinical outcomes by multivariable analysis.

Conclusions

Based on the pretreatment RDW and MSKCC scores, a novel predictive tool was established to stratify PCNS-DLBCL patients with different prognosis effectively. The predictive model developed accordingly is promising to judge the response of PCNS-DLBCL to methotrexate-based combination immunochemotherapy treatment. Thus, hematologists and oncologists could tailor and adjust therapeutic modalities by monitoring RDW in a prospective rather than the reactive manner, which could save medical expenditures and is a key concept in 3PM. In brief, RDW combined with MSKCC model could serve as an important tool for predicting the response to different treatment and the clinical outcomes for PCNS-DLBCL, which could conform with the principles of predictive, preventive, and personalized medicine.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13167-022-00290-5.

A novel machine learning model and a public online prediction platform for prediction of post-ERCP-cholecystitis (PEC)

BACKGROUND

Endoscopic retrograde cholangiopancreatography (ERCP) is an established treatment for common bile duct (CBD) stones. Post- ERCP cholecystitis (PEC) is a known complication of such procedure and there are no effective models and clinical applicable tools for PEC prediction.

METHODS

A random forest (RF) machine learning model was developed to predict PEC. Eligible patients at The First Hospital of Lanzhou University in China with common bile duct (CBD) stones and gallbladders in-situ were enrolled from 2010 to 2019. Logistic regression analysis was used to compare the predictive discrimination and accuracy values based on receiver operation characteristics (ROC) curve and decision and clinical impact curve. The RF model was further validated by another 117 patients. This study was registered with ClinicalTrials.gov, NCT04234126.

FINDINGS

A total of 1117 patients were enrolled (90 PEC, 8.06%) to build the predictive model for PEC. The RF method identified white blood cell (WBC) count, endoscopic papillary balloon dilatation (EPBD), increase in WBC, residual CBD stones after ERCP, serum amylase levels, and mechanical lithotripsy as the top six predictive factors and has a sensitivity of 0.822, specificity of 0.853 and accuracy of 0.855, with the area under curve (AUC) value of 0.890. A separate logistic regression prediction model was built with sensitivity, specificity, and AUC of 0.811, 0.791, and 0.864, respectively. An additional 117 patients (11 PEC, 9.40%) were used to validate the RF model, with an AUC of 0.889 compared to an AUC of 0.884 with the logistic regression model.

INTERPRETATION

The results suggest that the proposed RF model based on the top six PEC risk factors could be a promising tool to predict the occurrence of PEC.

Research Progress in the Early Warning of Chicken Diseases by Monitoring Clinical Symptoms

Global animal protein consumption has been steadily increasing as a result of population growth and the increasing demand for nutritious diets. The poultry industry provides a large portion of meat and eggs for human consumption. The early detection and warning of poultry infectious diseases play a critical role in the poultry breeding and production systems, improving animal welfare and reducing losses. However, inadequate methods for the early detection and prevention of infectious diseases in poultry farms sometimes fail to prevent decreased productivity and even widespread mortality. The health status of poultry is often reflected by its individual physiological, physical and behavioral clinical symptoms, such as higher body temperature resulting from fever, abnormal vocalization caused by respiratory disease and abnormal behaviors due to pathogenic infection. Therefore, the use of technologies for symptom detection can monitor the health status of broilers and laying hens in a continuous, noninvasive and automated way, and potentially assist in the early warning decision-making process. This review summarized recent literature on poultry disease detection and highlighted clinical symptom-monitoring technologies for sick poultry. The review concluded that current technologies are already showing their superiority to manual inspection, but the clinical symptom-based monitoring systems have not been fully utilized for on-farm early detection.

ABCModeller: an automatic data mining tool based on a consistent voting method with a user-friendly graphical interface

In order to extract useful information from a huge amount of biological data nowadays, simple and convenient tools are urgently needed for data analysis and modeling. In this paper, an automatic data mining tool, termed as ABCModeller (Automatic Binary Classification Modeller), with a user-friendly graphical interface was developed here, which includes automated functions as data preprocessing, significant feature extraction, classification modeling, model evaluation and prediction. In order to enhance the generalization ability of the final model, a consistent voting method was built here in this tool with the utilization of three popular machine-learning algorithms, as artificial neural network, support vector machine and random forest. Besides, Fibonacci search and orthogonal experimental design methods were also employed here to automatically select significant features in the data space and optimal hyperparameters of the three algorithms to achieve the best model. The reliability of this tool has been verified through multiple benchmark data sets. In addition, with the advantage of a user-friendly graphical interface of this tool, users without any programming skills can easily obtain reliable models directly from original data, which can reduce the complexity of modeling and data mining, and contribute to the development of related research including but not limited to biology. The excitable file of this tool can be downloaded from http://lishuyan.lzu.edu.cn/ABCModeller.rar.