Machine learning prediction model for post- hepatectomy liver failure in hepatocellular carcinoma: A multicenter study

Interpretable machine learning model for predicting post-hepatectomy liver failure in hepatocellular carcinoma

Post-hepatectomy liver failure (PHLF) is a severe complication following liver surgery. We aimed to develop a novel, interpretable machine learning (ML) model to predict PHLF. We enrolled 312 hepatocellular carcinoma (HCC) patients who underwent hepatectomy, and 30% of the samples were utilized for internal validation. Variable selection was performed using the least absolute shrinkage and selection operator regression in conjunction with random forest and recursive feature elimination (RF-RFE) algorithms. Subsequently, 12 distinct ML algorithms were employed to identify the optimal prediction model. The area under the receiver operating characteristic curve, calibration curves, and decision curve analysis (DCA) were utilized to assess the model's predictive accuracy. Additionally, an independent prospective validation was conducted with 62 patients. The SHapley Additive exPlanations (SHAP) analysis further explained the extreme gradient boosting (XGBoost) model. The XGBoost model exhibited the highest accuracy with AUCs of 0.983 and 0.981 in the training and validation cohorts among 12 ML models. Calibration curves and DCA confirmed the model's accuracy and clinical applicability. Compared with traditional models, the XGBoost model had a higher AUC. The prospective cohort (AUC = 0.942) further confirmed the generalization ability of the XGBoost model. SHAP identified the top three critical variables: total bilirubin (TBIL), MELD score, and ICG-R15. Moreover, the SHAP summary plot was used to illustrate the positive or negative effects of the features as influenced by XGBoost. The XGBoost model provides a good preoperative prediction of PHLF in patients with resectable HCC.

Multivariable prognostic models for post-hepatectomy liver failure: An updated systematic review

BACKGROUND

Partial hepatectomy continues to be the primary treatment approach for liver tumors, and post-hepatectomy liver failure (PHLF) remains the most critical life-threatening complication following surgery.

AIM

To comprehensively review the PHLF prognostic models developed in recent years and objectively assess the risk of bias in these models.

METHODS

This review followed the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guideline. Three databases were searched from November 2019 to December 2022, and references as well as cited literature in all included studies were manually screened in March 2023. Based on the defined inclusion criteria, articles on PHLF prognostic models were selected, and data from all included articles were extracted by two independent reviewers. The PROBAST was used to evaluate the quality of each included article.

RESULTS

A total of thirty-four studies met the eligibility criteria and were included in the analysis. Nearly all of the models (32/34, 94.1%) were developed and validated exclusively using private data sources. Predictive variables were categorized into five distinct types, with the majority of studies (32/34, 94.1%) utilizing multiple types of data. The area under the curve for the training models included ranged from 0.697 to 0.956. Analytical issues resulted in a high risk of bias across all studies included.

CONCLUSION

The validation performance of the existing models was substantially lower compared to the development models. All included studies were evaluated as having a high risk of bias, primarily due to issues within the analytical domain. The progression of modeling technology, particularly in artificial intelligence modeling, necessitates the use of suitable quality assessment tools.

Controlling nutritional status score predicts posthepatectomy liver failure: an online interpretable machine learning prediction model

BACKGROUND AND AIMS

Posthepatectomy liver failure (PHLF) remains a severe complication after hepatectomy for hepatocellular carcinoma (HCC) and accurate preoperative evaluation and predictive measures are urgently needed. We investigated the impact of the controlling nutritional status (CONUT) score on PHLF and utilized machine learning (ML) algorithms to identify high-risk individuals of PHLF.

METHOD

A total of 464 patients with HCC undergoing hepatectomy were randomized 7 : 2: 1 into the training group (n = 324), test group (n = 96), and validation group (n = 46). In the training group, variables were screened by univariate logistic regression combined with least absolute shrinkage and selection operator regression. Models were then developed using nine ML algorithms and the optimal model was interpreted via SHapley Additive exPlanations and deployed online.

RESULTS

PHLF was present in 29 of 324 (8.9%) patients. The light gradient boosting machine (LightGBM) model based on the CONUT score exhibited excellent performance, with an area under the curve (AUC) of 0.927 [95% confidence interval (CI): 0.886-0.967], an area under the precision-recall curve (AUPRC) of 0.644 (95% CI: 0.469-0.785), and a Brier score of 0.055 in the training group. And an AUC of 0.703 (95% CI: 0.528-0.879), an AUPRC of 0.420 (95% CI: 0.096-0.703), and a Brier score of 0.091 in the test group. In the validation group, AUC, AUPRC, and Brier score were 0.808 (95% CI: 0.637-0.980), 0.516 (95% CI: 0.086-0.841), and 0.096, respectively. The model was made available online for clinical application (LightGBM for PHLF).

CONCLUSION

The CONUT score significantly influences PHLF. The LightGBM model demonstrates the prominent predictive capacity of PHLF.

Utility of a novel scoring system for difficulty of pure laparoscopic hepatectomy for intrahepatic cholangiocarcinoma

Despite the growing adoption of laparoscopic hepatectomy (LH) for intrahepatic cholangiocarcinoma (ICC), there is no scoring system available designed to evaluate its surgical complexity. This paper aims to introduce a novel difficulty scoring system (DSS), designated as the Wei-DSS, exclusively tailored to assess the surgical difficulty of pure LH for ICC. We retrospectively collected clinical data from ICC patients who underwent pure LH at our institution, spanning from November 2018 to May 2024. Patients were categorized into two levels of Wei-DSS scores (low-difficulty [5-6], and high-difficulty [7-10]) determined by tumor characteristics, liver texture, resection extent and tumor marker levels. A total of 104 patients were enrolled in this study including a low-difficulty (LD) group comprising 47 patients and a high-difficulty (HD) group comprising 57 patients. Perioperative comparisons indicated that the HD group was significantly associated with a longer operation time (318.14 ± 125.89 min vs. 222.83 ± 119.03 min, P < 0.001), higher rates of intraoperative blood transfusions (59.6% vs. 27.7%, P = 0.001), and increased rates of postoperative complications (84.2% vs. 48.9%, P < 0.001) compared to the LD group. The receiver operating characteristic (ROC) curve analysis indicated that the Wei-DSS demonstrated superior predictive accuracy over the Major/Minor Classification for predicting postoperative complication rates (area under the curve [AUC] 0.702 vs. 0.622) and operating time (AUC 0.720 vs. 0.604 ). The Wei-DSS score may have the potential to assist surgeons in categorizing ICC patients with varying levels of surgical difficulty of LH, though it warrants further validations across multiple centers to solidify its efficacy and reliability.

Comparison of multiple machine learning models for predicting prognosis of pancreatic ductal adenocarcinoma based on contrast-enhanced CT radiomics and clinical features

Objective

The aim of this study was to evaluate the prognostic potential of combining clinical features and radiomics with multiple machine learning (ML) algorithms in pancreatic ductal adenocarcinoma (PDAC).

Methods

A total of 116 patients with PDAC who met the eligibility criteria were randomly assigned to a training or validation cohort. Seven ML algorithms, including Supervised Principal Components, stepwise Cox, Random Survival Forest, CoxBoost, Least absolute shrinkage and selection operation (Lasso), Ridge, and Elastic network, were integrated into 43 algorithm combinations. Forty-three radiomics models were constructed separately using radiomics features extracted from arterial phase (AP), venous phase (VP), and combined arterial and venous phase (AP+VP) images. The concordance index (C-index) of each model was calculated. The model with the highest mean C-index was identified as the best model for calculating the radiomics score (Radscore). Univariate and multivariate Cox analyses were used to identify independent prognostic indicators and create a clinical model for prognosis prediction. The multivariable Cox regression was used to combine Radscore with clinical features to create a combined model. The efficacy of the model was evaluated using the C-index, calibration curves, and decision curve analysis (DCA).

Results

The model based on the Lasso+StepCox[both] algorithm constructed using AP+VP radiomics features showed the best predictive ability among the 114 radiomics models. The C-indices of the model in the training and validation cohorts were 0.742 and 0.722, respectively. Based on the results of the univariate and multivariate Cox regression analyses, sex, Tumor-Node-Metastasis (TNM) stage, and systemic inflammation response index were included to build the clinical model. The combined model, incorporating three clinical factors and AP+VP-Radscore, achieved the highest C-indices of 0.764 and 0.746 in the training and validation cohorts, respectively. In terms of preoperative prognosis prediction for PDAC, the calibration curve and DCA showed that the combined model had a good consistency and greatest net benefit.

Conclusion

A combined model of clinical features and AP+VP-Radscore screened using multiple ML algorithms has an excellent ability to predict the prognosis of PDAC and may provide a noninvasive and effective method for clinical decision-making.

Online interpretable dynamic prediction models for clinically significant posthepatectomy liver failure based on machine learning algorithms: a retrospective cohort study

BACKGROUND

Posthepatectomy liver failure (PHLF) is the leading cause of mortality in patients undergoing hepatectomy. However, practical models for accurately predicting the risk of PHLF are lacking. This study aimed to develop precise prediction models for clinically significant PHLF.

METHODS

A total of 226 patients undergoing hepatectomy at a single center were recruited. The study outcome was clinically significant PHLF. Five preoperative and postoperative machine learning (ML) models were developed and compared with four clinical scores, namely, the MELD, FIB-4, ALBI, and APRI scores. The robustness of the developed ML models was internally validated using fivefold cross-validation (CV) by calculating the average of the evaluation metrics and was externally validated on an independent temporal dataset, including the area under the curve (AUC) and the area under the precision-recall curve (AUPRC). SHapley Additive exPlanations analysis was performed to interpret the best performance model.

RESULTS

Clinically significant PHLF was observed in 23 of 226 patients (10.2%). The variables in the preoperative model included creatinine, total bilirubin, and Child-Pugh grade. In addition to the above factors, the extent of resection was also a key variable for the postoperative model. The preoperative and postoperative artificial neural network (ANN) models exhibited excellent performance, with mean AUCs of 0.766 and 0.851, respectively, and mean AUPRC values of 0.441 and 0.645, whereas the MELD, FIB-4, ALBI, and APRI scores reached AUCs of 0.714, 0.498, 0.536, and 0.551, respectively, and AUPRC values of 0.204, 0.111, 0.128, and 0.163, respectively. In addition, the AUCs of the preoperative and postoperative ANN models were 0.720 and 0.731, respectively, and the AUPRC values were 0.380 and 0.408, respectively, on the temporal dataset.

CONCLUSION

Our online interpretable dynamic ML models outperformed common clinical scores and could function as a clinical decision support tool to identify patients at high risk of PHLF preoperatively and postoperatively.

A Real-World Study on the Short-Term Efficacy of Amlodipine in Treating Hypertension Among Inpatients

Purpose

Hospitalized hypertensive patients rely on blood pressure medication, yet there is limited research on the sole use of amlodipine, despite its proven efficacy in protecting target organs and reducing mortality. This study aims to identify key indicators influencing the efficacy of amlodipine, thereby enhancing treatment outcomes.

Patients and Methods

In this multicenter retrospective study, 870 hospitalized patients with primary hypertension exclusively received amlodipine for the first 5 days after admission, and their medical records contained comprehensive blood pressure records. They were categorized into success (n=479) and failure (n=391) groups based on average blood pressure control efficacy. Predictive models were constructed using six machine learning algorithms. Evaluation metrics encompassed the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). SHapley Additive exPlanations (SHAP) analysis assessed feature contributions to efficacy.

Results

All six machine learning models demonstrated superior predictive performance. Following variable reduction, the model predicting amlodipine efficacy was reconstructed using these algorithms, with the light gradient boosting machine (LightGBM) model achieving the highest overall performance (AUC = 0.803). Notably, amlodipine showed enhanced efficacy in patients with low platelet distribution width (PDW) values, as well as high hematocrit (HCT) and thrombin time (TT) values.

Conclusion

This study utilized machine learning to predict amlodipine's effectiveness in hypertension treatment, pinpointing key factors: HCT, PDW, and TT levels. Lower PDW, along with higher HCT and TT, correlated with enhanced treatment outcomes. This facilitates personalized treatment, particularly for hospitalized hypertensive patients undergoing amlodipine monotherapy.

Utility of Machine Learning in the Prediction of Post-Hepatectomy Liver Failure in Liver Cancer

Background

Posthepatectomy liver failure (PHLF) is a serious complication associated with high mortality rates. Machine learning (ML) has rapidly developed and may outperform traditional models in predicting PHLF in patients who have undergone hepatectomy. This study aimed to predict PHLF using ML and compare its performance with that of traditional scoring systems.

Methods

The clinicopathological data of 334 patients who underwent liver resection were retrospectively collected. The Pycaret library, a simple, open-source machine learning library, was used to compare multiple classification models for PHLF prediction. The predictive performance of 15 ML algorithms was compared using the mean area under the receiver operating characteristic curve (AUROC) and accuracy, and the best-fit model was selected among 15 ML algorithms. Next, the predictive performance of the selected ML-PHLF model was compared with that of routine scoring systems, the albumin-bilirubin score (ALBI) and the fibrosis-4 (FIB-4) index, using AUROC.

Results

The best model was extreme gradient boosting (accuracy:93.1%; AUROC:0.863) among the 15 ML algorithms. As compared with ALBI and FIB-4, the ML PHLF model had higher AUROC for predicting PHLF.

Conclusion

The novel ML model for predicting PHLF outperformed routine scoring systems.

Multi-omic validation of the cuproptosis-sphingolipid metabolism network: modulating the immune landscape in osteosarcoma

Background

The current understanding of the mechanisms by which metal ion metabolism promotes the progression and drug resistance of osteosarcoma remains incomplete. This study aims to elucidate the key roles and mechanisms of genes involved in cuproptosis-related sphingolipid metabolism (cuproptosis-SPGs) in regulating the immune landscape, tumor metastasis, and drug resistance in osteosarcoma cells.

Methods

This study employed multi-omics approaches to assess the impact of cuproptosis-SPGs on the prognosis of osteosarcoma patients. Lasso regression analysis was utilized to construct a prognostic model, while multivariate regression analysis was applied to identify key core genes and generate risk coefficients for these genes, thereby calculating a risk score for each osteosarcoma patient. Patients were then stratified into high-risk and low-risk groups based on their risk scores. The ESTIMATE and CIBERSORT algorithms were used to analyze the level of immune cell infiltration within these risk groups to construct the immune landscape. Single-cell analysis was conducted to provide a more precise depiction of the expression patterns of cuproptosis-SPGs among immune cell subtypes. Finally, experiments on osteosarcoma cells were performed to validate the role of the cuproptosis-sphingolipid signaling network in regulating cell migration and apoptosis.

Results

In this study, seven cuproptosis-SPGs were identified and used to construct a prognostic model for osteosarcoma patients. In addition to predicting survival, the model also demonstrated reliability in forecasting the response to chemotherapy drugs. The results showed that a high cuproptosis-sphingolipid metabolism score was closely associated with reduced CD8 T cell infiltration and indicated poor prognosis in osteosarcoma patients. Cellular functional assays revealed that cuproptosis-SPGs regulated the LC3B/ERK signaling pathway, thereby triggering cell death and impairing migration capabilities in osteosarcoma cells.

Conclusion

The impact of cuproptosis-related sphingolipid metabolism on the survival and migration of osteosarcoma cells, as well as on CD8 T cell infiltration, highlights the potential of targeting copper ion metabolism as a promising strategy for osteosarcoma patients.

A machine learning model based on clinical features and ultrasound radiomics features for pancreatic tumor classification

Objective

This study aimed to construct a machine learning model using clinical variables and ultrasound radiomics features for the prediction of the benign or malignant nature of pancreatic tumors.

Methods

242 pancreatic tumor patients who were hospitalized at the First Affiliated Hospital of Guangxi Medical University between January 2020 and June 2023 were included in this retrospective study. The patients were randomly divided into a training cohort (n=169) and a test cohort (n=73). We collected 28 clinical features from the patients. Concurrently, 306 radiomics features were extracted from the ultrasound images of the patients' tumors. Initially, a clinical model was constructed using the logistic regression algorithm. Subsequently, radiomics models were built using SVM, random forest, XGBoost, and KNN algorithms. Finally, we combined clinical features with a new feature RAD prob calculated by applying radiomics model to construct a fusion model, and developed a nomogram based on the fusion model.

Results

The performance of the fusion model surpassed that of both the clinical and radiomics models. In the training cohort, the fusion model achieved an AUC of 0.978 (95% CI: 0.96-0.99) during 5-fold cross-validation and an AUC of 0.925 (95% CI: 0.86-0.98) in the test cohort. Calibration curve and decision curve analyses demonstrated that the nomogram constructed from the fusion model has high accuracy and clinical utility.

Conclusion

The fusion model containing clinical and ultrasound radiomics features showed excellent performance in predicting the benign or malignant nature of pancreatic tumors.

Multi-omics analysis and experimental validation of the value of monocyte-associated features in prostate cancer prognosis and immunotherapy

Background

Monocytes play a critical role in tumor initiation and progression, with their impact on prostate adenocarcinoma (PRAD) not yet fully understood. This study aimed to identify key monocyte-related genes and elucidate their mechanisms in PRAD.

Method

Utilizing the TCGA-PRAD dataset, immune cell infiltration levels were assessed using CIBERSORT, and their correlation with patient prognosis was analyzed. The WGCNA method pinpointed 14 crucial monocyte-related genes. A diagnostic model focused on monocytes was developed using a combination of machine learning algorithms, while a prognostic model was created using the LASSO algorithm, both of which were validated. Random forest and gradient boosting machine singled out CCNA2 as the most significant gene related to prognosis in monocytes, with its function further investigated through gene enrichment analysis. Mendelian randomization analysis of the association of HLA-DR high-expressing monocytes with PRAD. Molecular docking was employed to assess the binding affinity of CCNA2 with targeted drugs for PRAD, and experimental validation confirmed the expression and prognostic value of CCNA2 in PRAD.

Result

Based on the identification of 14 monocyte-related genes by WGCNA, we developed a diagnostic model for PRAD using a combination of multiple machine learning algorithms. Additionally, we constructed a prognostic model using the LASSO algorithm, both of which demonstrated excellent predictive capabilities. Analysis with random forest and gradient boosting machine algorithms further supported the potential prognostic value of CCNA2 in PRAD. Gene enrichment analysis revealed the association of CCNA2 with the regulation of cell cycle and cellular senescence in PRAD. Mendelian randomization analysis confirmed that monocytes expressing high levels of HLA-DR may promote PRAD. Molecular docking results suggested a strong affinity of CCNA2 for drugs targeting PRAD. Furthermore, immunohistochemistry experiments validated the upregulation of CCNA2 expression in PRAD and its correlation with patient prognosis.

Conclusion

Our findings offer new insights into monocyte heterogeneity and its role in PRAD. Furthermore, CCNA2 holds potential as a novel targeted drug for PRAD.

Artificial Intelligence in Perioperative Planning and Management of Liver Resection

Artificial intelligence (AI) is a speciality within computer science that deals with creating systems that can replicate the intelligence of a human mind and has problem-solving abilities. AI includes a diverse array of techniques and approaches such as machine learning, neural networks, natural language processing, robotics, and expert systems. An electronic literature search was conducted using the databases of "PubMed" and "Google Scholar". The period for the search was from 2000 to June 2023. The search terms included "artificial intelligence", "machine learning", "liver cancers", "liver tumors", "hepatectomy", "perioperative" and their synonyms in various combinations. The search also included all MeSH terms. The extracted articles were further reviewed in a step-wise manner for identification of relevant studies. A total of 148 articles were identified after the initial literature search. Initial review included screening of article titles for relevance and identifying duplicates. Finally, 65 articles were reviewed for this review article. The future of AI in liver cancer planning and management holds immense promise. AI-driven advancements will increasingly enable precise tumour detection, location, and characterisation through enhanced image analysis. ML algorithms will predict patient-specific treatment responses and complications, allowing for tailored therapies. Surgical robots and AI-guided procedures will enhance the precision of liver resections, reducing risks and improving outcomes. AI will also streamline patient monitoring, better hemodynamic management, enabling early detection of recurrence or complications. Moreover, AI will facilitate data-driven research, accelerating the development of novel treatments and therapies. Ultimately, AI's integration will revolutionise liver cancer care, offering personalised, efficient and effective solutions, improving patients' quality of life and survival rates.

Explainable artificial intelligence prediction-based model in laparoscopic liver surgery for segments 7 and 8: an international multicenter study

BACKGROUND

Artificial intelligence (AI) is becoming more useful as a decision-making and outcomes predictor tool. We have developed AI models to predict surgical complexity and the postoperative course in laparoscopic liver surgery for segments 7 and 8.

METHODS

We included patients with lesions located in segments 7 and 8 operated by minimally invasive liver surgery from an international multi-institutional database. We have employed AI models to predict surgical complexity and postoperative outcomes. Furthermore, we have applied SHapley Additive exPlanations (SHAP) to make the AI models interpretable. Finally, we analyzed the surgeries not converted to open versus those converted to open.

RESULTS

Overall, 585 patients and 22 variables were included. Multi-layer Perceptron (MLP) showed the highest performance for predicting surgery complexity and Random Forest (RF) for predicting postoperative outcomes. SHAP detected that MLP and RF gave the highest relevance to the variables "resection type" and "largest tumor size" for predicting surgery complexity and postoperative outcomes. In addition, we explored between surgeries converted to open and non-converted, finding statistically significant differences in the variables "tumor location," "blood loss," "complications," and "operation time."

CONCLUSION

We have observed how the application of SHAP allows us to understand the predictions of AI models in surgical complexity and the postoperative outcomes of laparoscopic liver surgery in segments 7 and 8.

The Use of Artificial Intelligence in the Liver Histopathology Field: A Systematic Review

Digital pathology (DP) has begun to play a key role in the evaluation of liver specimens. Recent studies have shown that a workflow that combines DP and artificial intelligence (AI) applied to histopathology has potential value in supporting the diagnosis, treatment evaluation, and prognosis prediction of liver diseases. Here, we provide a systematic review of the use of this workflow in the field of hepatology. Based on the PRISMA 2020 criteria, a search of the PubMed, SCOPUS, and Embase electronic databases was conducted, applying inclusion/exclusion filters. The articles were evaluated by two independent reviewers, who extracted the specifications and objectives of each study, the AI tools used, and the results obtained. From the 266 initial records identified, 25 eligible studies were selected, mainly conducted on human liver tissues. Most of the studies were performed using whole-slide imaging systems for imaging acquisition and applying different machine learning and deep learning methods for image pre-processing, segmentation, feature extractions, and classification. Of note, most of the studies selected demonstrated good performance as classifiers of liver histological images compared to pathologist annotations. Promising results to date bode well for the not-too-distant inclusion of these techniques in clinical practice.

Predicting Safe Liver Resection Volume for Major Hepatectomy Using Artificial Intelligence

(1) Background: Advancements in the field of liver surgery have led to a critical need for precise estimations of preoperative liver function to prevent post-hepatectomy liver failure (PHLF), a significant cause of morbidity and mortality. This study introduces a novel application of artificial intelligence (AI) in determining safe resection volumes according to a patient's liver function in major hepatectomies. (2) Methods: We incorporated a deep learning approach, incorporating a unique liver-specific loss function, to analyze patient characteristics, laboratory data, and liver volumetry from computed tomography scans of 52 patients. Our approach was evaluated against existing machine and deep learning techniques. (3) Results: Our approach achieved 68.8% accuracy in predicting safe resection volumes, demonstrating superior performance over traditional models. Furthermore, it significantly reduced the mean absolute error in under-predicted volumes to 23.72, indicating a more precise estimation of safe resection limits. These findings highlight the potential of integrating AI into surgical planning for liver resections. (4) Conclusion: By providing more accurate predictions of safe resection volumes, our method aims to minimize the risk of PHLF, thereby improving clinical outcomes for patients undergoing hepatectomy.