A New Approach for Determining Short-Term, Objective Prognostic Predictive Methods for Terminal Cancer Patients Based on the Change Point of Laboratory Test Values

Development and Validation of an Interpretable Machine Learning Model for Early Prognosis Prediction in ICU Patients with Malignant Tumors and Hyperkalemia

This study aims to develop and validate a machine learning (ML) predictive model for assessing mortality in patients with malignant tumors and hyperkalemia (MTH). We extracted data on patients with MTH from the Medical Information Mart for Intensive Care-IV, version 2.2 (MIMIC-IV v2.2) database. The dataset was split into a training set (75%) and a validation set (25%). We used the Least Absolute Shrinkage and Selection Operator (LASSO) regression to identify potential predictors, which included clinical laboratory indicators and vital signs. Pearson correlation analysis tested the correlation between predictors. In-hospital death was the prediction target. The Area Under the Curve (AUC) and accuracy of the training and validation sets of 7 ML algorithms were compared, and the optimal 1 was selected to develop the model. The calibration curve was used to evaluate the prediction accuracy of the model further. SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) enhanced model interpretability. 496 patients with MTH in the Intensive Care Unit (ICU) were included. After screening, 17 clinical features were included in the construction of the ML model, and the Pearson correlation coefficient was <0.8, indicating that the correlation between the clinical features was small. eXtreme Gradient Boosting (XGBoost) outperformed other algorithms, achieving perfect scores in the training set (accuracy: 1.000, AUC: 1.000) and high scores in the validation set (accuracy: 0.734, AUC: 0.733). The calibration curves indicated good predictive calibration of the model. SHAP analysis identified the top 8 predictive factors: urine output, mean heart rate, maximum urea nitrogen, minimum oxygen saturation, minimum mean blood pressure, maximum total bilirubin, mean respiratory rate, and minimum pH. In addition, SHAP and LIME performed in-depth individual case analyses. This study demonstrates the effectiveness of ML methods in predicting mortality risk in ICU patients with MTH. It highlights the importance of predictors like urine output and mean heart rate. SHAP and LIME significantly enhanced the model's interpretability.

Prognostic model for patients with advanced cancer using a combination of routine blood test values

PURPOSE

The purpose of this study was to develop a simple prognostic model based on objective indicators alone, i.e., routine blood test data, without using any subjective variables such as patient's symptoms and physician's prediction.

METHODS

The subjects of this retrospective study were patients at the palliative care unit of Tohoku University Hospital, Japan. Eligible patients were over 20 years old and had advanced cancer (n = 225). The model for predicting survival was developed based on Cox proportional hazards regression models for univariable and multivariable analyses of 20 items selected from routine blood test data. All the analyses were performed according to the TRIPOD statement ( https://www.tripod-statement.org/ ).

RESULTS

The univariable and multivariable regression analyses identified total bilirubin, creatinine, urea/creatinine ratio, aspartate aminotransferase, albumin, total leukocyte count, differential lymphocyte count, and platelet/lymphocyte ratio as significant risk factors for mortality. Based on the hazard ratios, the area under the curve for the new risk model was 0.87 for accuracy, 0.83 for sensitivity, and 0.74 for specificity. Diagnostic accuracy was higher than provided by the Palliative Prognostic Score and the Palliative Prognostic Index. The Kaplan-Meier analysis demonstrated a survival significance of classifying patients according to their score into low-, medium-, and high-mortality risk groups having median survival times of 67 days, 34 days, and 11 days, respectively (p < 0.001).

CONCLUSIONS

We developed a simple and accurate prognostic model for predicting the survival of patients with advanced cancer based on routine blood test values alone that may be useful for appropriate advanced care planning in a palliative care setting.

[Global Standards for Pharmaceutical Education]

The environment surrounding clinical pharmacy practices has changed greatly in the past thirty-some years, basically since the end of the 1980s. During this period, the separation ratio between pharmacists' dispensing and prescribing functions has increased, from 12% to 74%. The three big events in this timeline include the beginning of pharmaceutical care for inpatients by hospital pharmacists in 1988; the transition of pharmacy schools to a six-year educational program in 2006; and the revision of Pharmaceutical Affairs Law, as well as its name change, in 2014. In concert with these events, the central role of the pharmacist has changed from being dispensing-centric to an active participation in patient treatment via medication as a member of the medical care team. As a key participant in these changes, the author helped to improve the operations of hospital pharmacists, strengthened their role with advanced information and communication technology (ICT) support, and established a baseline for clinical pharmacy research and education. Accordingly, in this paper, the history of this development will be reviewed, and the future of a global standard for pharmaceutical education will be discussed.

Validation of a Short-Term, Objective, Prognostic Predictive Method for Terminal Cancer Patients in a Palliative Care Unit Using a Combination of Six Laboratory Test Items

Background: There is no established method to objectively predict short-term prognosis. Recently, we proposed objective, short-term, prognostic predictive methods that are combinations of laboratory test items: WPCBAL score, derived from six values (white blood cell, platelet, C-reactive protein, blood urea nitrogen, aspartate aminotransferase, and lactate dehydrogenase). However, that study was conducted in an acute-phase hospital to identify the test items useful for prognostic prediction; thus, whether WPCBAL score could be applied to terminal cancer patients in a palliative care unit was unverified. Objective: To verify the usefulness of WPCBAL score for terminal cancer patients. Design: A retrospective study. Setting/Subjects: Patients admitted to the palliative care unit of Ashiya Municipal Hospital (N = 128) in Japan in 2016. Measurements: The sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and area under the receiver operating characteristic curve (AUROC) were compared between WPCBAL score and the Glasgow prognostic score (GPS). Results: For predicting three-week prognosis, WPCBAL score showed higher AUROC compared with GPS (0.7540 and 0.6573, respectively). WPCBAL score predicting two-week prognosis showed greater AUROC than GPS predicting three-week prognosis (0.7491 and 0.6573, respectively). Conclusion: WPCBAL score was verified to objectively predict the two- or three-week prognosis for terminal cancer patients in a palliative care unit. WPCBAL score may be a new option for prognostic prediction for terminal cancer patients.