This work introduces a predictive Length of Stay (LOS) framework for lung cancer patients using machine learning (ML) models. The framework proposed to deal with imbalanced datasets for classification-based approaches using electronic healthcare records (EHR). We have utilized supervised ML methods to predict lung cancer inpatients LOS during ICU hospitalization using the MIMIC-III dataset. Random Forest (RF) Model outperformed other models and achieved predicted results during the three framework phases. With clinical significance features selection, over-sampling methods (SMOTE and ADASYN) achieved the highest AUC results (98% with CI 95%: 95.3–100%, and 100% respectively). The combination of Over-sampling and under-sampling achieved the second-highest AUC results (98%, with CI 95%: 95.3–100%, and 97%, CI 95%: 93.7–100% SMOTE-Tomek, and SMOTE-ENN respectively). Under-sampling methods reported the least important AUC results (50%, with CI 95%: 40.2–59.8%) for both (ENN and Tomek- Links). Using ML explainable technique called SHAP, we explained the outcome of the predictive model (RF) with SMOTE class balancing technique to understand the most significant clinical features that contributed to predicting lung cancer LOS with the RF model. Our promising framework allows us to employ ML techniques in-hospital clinical information systems to predict lung cancer admissions into ICU.
Biliary tract cancers (BTC) represent an aggressive disease with a dismal prognosis. Gemcitabine in combination with cisplatin is the standard first-line palliative treatment for advanced BTC. There is no established treatment following progression on gemcitabine-cisplatin. In this article, we present two cases for individuals with advanced BTC who were treated with pembrolizumab and the tumors have completely resolved.
Objective: Chemotherapy-induced febrile neutropenia is a common and serious oncological emergency which carries a substantial mortality and morbidity. The main objective of this study is to evaluate the usage of absolute monocyte count (AMC) at presentation as a prognostic factor for patients with chemotherapy-induced febrile neutropenia who were subsequently treated with granulocyte colony-stimulating factor (G-CSF). Study Design: The electronic medical records of our center were used retrospectively to identify patients diagnosed with unprecedented chemotherapy-induced febrile neutropenia treated with G-CSF between January 2010 to December 2020 and diagnosed with solid and hematological malignancies. Patient's demographics, disease characteristics and laboratory investigations were extracted. Disease progression measures were statistically compared between the study groups in the short-term period of follow-up (six days) including absolute neutrophil count (ANC), ANC difference compared to the baseline readings, hospitalization period, and mortality. Results: A total of 80 patients were identified and categorized into two groups namely monocytopenia (n = 34) and non-monocytopenia (n = 46) with an AMC cutoff point of 0.1×10 9 cells/L. The monocytopenia group exhibited a worse prognosis with lower ANC values and slower improvement illustrated by the low ANC difference values at all follow up points (P-value ≤ 0.05) apart from day 5. A statistically significant lower hospitalization period was also observed in the non-monocytopenia group (P-value = 0.006). Linear regression analysis evaluated the association between AMC values at admission and ANC values at admission along with subsequent days of follow up which were found to be statistically significant (P-value ≤ 0.05). Receiver operating characteristic curves suggest a satisfactory predictability of ANC changes by AMC values at admission, days1, 2, 3, 4 and 6. Conclusion: Monocytopenia holds a worse prognosis in chemotherapy-induced febrile neutropenia patients treated with G-CSF. In addition, AMC values at presentation represents a potential risk factor that can predict short-term changes regarding ANC measures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.