Dissecting bloodstream infections in febrile neutropenic patients with hematological malignancies, a decade-long single center retrospective observational study (2009-2019).

BACKGROUND: The use of ill-suited antibiotics is a significant risk factor behind the increase in the mortality, morbidity, and economic burden for patients who are under treatment for hematological malignancy (HM) and bloodstream infections (BSI). Such unfitting treatment choices intensify the evolution of resistant variants which is a public health concern due to possible healthcare-associated infection spread to the general population. Hence, this study aims to evaluate antibiograms of patients with BSI and risk factors associated with septicemia. METHODS: A total of 1166 febrile neutropenia episodes (FNE) among 513 patients with HM from the National Center for Cancer Care and Research (NCCCR), Qatar, during 2009-2019 were used for this study. The socio-demographic, clinical, microbial, and anti-microbial data retrieved from the patient's health records were used. RESULTS: We analyzed the sensitivity of gram-negative and gram-positive bacilli reported in HM-FN-BSI patients. Out of the total 512 microorganisms isolated, 416 (81%) were gram-negative bacteria (GNB), 76 (15%) were gram-positive bacteria (GPB) and 20 (4%) were fungi. Furthermore, in 416 GNB, 298 (71.6%) were Enterobacteriaceae sp. among which 121 (41%) were ESBL (Extended Spectrum Beta-Lactamase) resistant to Cephalosporine third generation and Piperacillin-Tazobactam, 54 (18%) were Carbapenem-resistant or multidrug-resistant organism (MDRO). It's noteworthy that the predominant infectious agents in our hospital include E. coli, Klebsiella species, and P. aeruginosa. Throughout the study period, the mortality rate due to BSI was 23%. Risk factors that show a significant correlation with death are age, disease status, mono or polymicrobial BSI and septic shock. CONCLUSION: Decision pertaining to the usage of antimicrobials for HM-FN-BSI patients is a critical task that relies on the latest pattern of prevalence, treatment resistance, and clinical outcomes. Analysis of the antibiogram of HM-FN-BSI patients in Qatar calls for a reconsideration of currently followed empirical antibiotic therapy towards better infection control and antimicrobial stewardship.

An Artificial Intelligence-Based Diagnostic System for Acute Lymphoblastic Leukemia Detection.

This study suggests a novel Acute Lymphoblastic Leukemia (ALL) diagnostic model, built solely on complete blood count (CBC) records. Using a dataset comprised of CBC records of 86 ALL and 86 control patients respectively, we identified the most ALL-specific parameters using a feature selection approach. Next, Grid Search-based hyperparameter tuning with a five-fold cross-validation scheme was adopted to build classifiers using Random Forest, XGBoost, and Decision Tree algorithms. A comparison between the performances of the three models demonstrates that Decision Tree classifier outperformed XGBoost and Random Forest algorithms in ALL detection using CBC-based records.

Machine Learning for Diagnosis and Screening of Chronic Lymphocytic Leukemia Using Routine Complete Blood Count (CBC) Results.

The comprehensive epidemiology and global disease burdens reported recently suggest that chronic lymphocytic leukemia (CLL) constitutes 25-30% of leukemias thus being the most common leukemia subtype. However, there is an insufficient presence of artificial intelligence (AI)-based techniques for CLL diagnosis. The novelty of this study is in the investigation of data-driven techniques to leverage the intricate CLL-related immune dysfunctions reflected in routine complete blood count (CBC) alone. We used statistical inferences, four feature selection methods, and multistage hyperparameter tuning to build robust classifiers. With respective accuracies of 97.05%, 97.63%, and 98.62% for Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), and XGboost (XGb)-based models, CBC-driven AI methods promise timely medical care and improved patient outcome with lesser resource usage and related cost.

Clustering and Stochastic Simulation Optimization for Outpatient Chemotherapy Appointment Planning and Scheduling.

Outpatient Chemotherapy Appointment (OCA) planning and scheduling is a process of distributing appointments to available days and times to be handled by various resources through a multi-stage process. Proper OCAs planning and scheduling results in minimizing the length of stay of patients and staff overtime. The integrated consideration of the available capacity, resources planning, scheduling policy, drug preparation requirements, and resources-to-patients assignment can improve the Outpatient Chemotherapy Process's (OCP's) overall performance due to interdependencies. However, developing a comprehensive and stochastic decision support system in the OCP environment is complex. Thus, the multi-stages of OCP, stochastic durations, probability of uncertain events occurrence, patterns of patient arrivals, acuity levels of nurses, demand variety, and complex patient pathways are rarely addressed together. Therefore, this paper proposes a clustering and stochastic optimization methodology to handle the various challenges of OCA planning and scheduling. A Stochastic Discrete Simulation-Based Multi-Objective Optimization (SDSMO) model is developed and linked to clustering algorithms using an iterative sequential approach. The experimental results indicate the positive effect of clustering similar appointments on the performance measures and the computational time. The developed cluster-based stochastic optimization approaches showed superior performance compared with baseline and sequencing heuristics using data from a real Outpatient Chemotherapy Center (OCC).

Quantification of the growth suppression of HER2+ breast cancer colonies under the effect of trastuzumab and PD-1/PD-L1 inhibitor.

Introduction: Immune checkpoint blockade (ICB)-based therapy is revolutionizing cancer treatment by fostering successful immune surveillance and effector cell responses against various types of cancers. However, patients with HER2+ cancers are yet to benefit from this therapeutic strategy. Precisely, several questions regarding the right combination of drugs, drug modality, and effective dose recommendations pertaining to the use of ICB-based therapy for HER2+ patients remain unanswered. Methods: In this study, we use a mathematical modeling-based approach to quantify the growth inhibition of HER2+ breast cancer (BC) cell colonies (ZR75) when treated with anti-HER2; trastuzumab (TZ) and anti-PD-1/PD-L1 (BMS-202) agents. Results and discussion: Our data show that a combination therapy of TZ and BMS-202 can significantly reduce the viability of ZR75 cells and trigger several morphological changes. The combination decreased the cell's invasiveness along with altering several key pathways, such as Akt/mTor and ErbB2 compared to monotherapy. In addition, BMS-202 causes dose-dependent growth inhibition of HER2+ BC cell colonies alone, while this effect is significantly improved when used in combination with TZ. Based on the in-vitro monoculture experiments conducted, we argue that BMS-202 can cause tumor growth suppression not only by mediating immune response but also by interfering with the growth signaling pathways of HER2+BC. Nevertheless, further studies are imperative to substantiate this argument and to uncover the potential crosstalk between PD-1/PD-L1 inhibitors and HER2 growth signaling pathways in breast cancer.

A Review of Artificial Intelligence Applications in Hematology Management: Current Practices and Future Prospects.

BACKGROUND: Machine learning (ML) and deep learning (DL) methods have recently garnered a great deal of attention in the field of cancer research by making a noticeable contribution to the growth of predictive medicine and modern oncological practices. Considerable focus has been particularly directed toward hematologic malignancies because of the complexity in detecting early symptoms. Many patients with blood cancer do not get properly diagnosed until their cancer has reached an advanced stage with limited treatment prospects. Hence, the state-of-the-art revolves around the latest artificial intelligence (AI) applications in hematology management. OBJECTIVE: This comprehensive review provides an in-depth analysis of the current AI practices in the field of hematology. Our objective is to explore the ML and DL applications in blood cancer research, with a special focus on the type of hematologic malignancies and the patient's cancer stage to determine future research directions in blood cancer. METHODS: We searched a set of recognized databases (Scopus, Springer, and Web of Science) using a selected number of keywords. We included studies written in English and published between 2015 and 2021. For each study, we identified the ML and DL techniques used and highlighted the performance of each model. RESULTS: Using the aforementioned inclusion criteria, the search resulted in 567 papers, of which 144 were selected for review. CONCLUSIONS: The current literature suggests that the application of AI in the field of hematology has generated impressive results in the screening, diagnosis, and treatment stages. Nevertheless, optimizing the patient's pathway to treatment requires a prior prediction of the malignancy based on the patient's symptoms or blood records, which is an area that has still not been properly investigated.

Prediction of Multiple Clinical Complications in Cancer Patients to Ensure Hospital Preparedness and Improved Cancer Care.

Reliable and rapid medical diagnosis is the cornerstone for improving the survival rate and quality of life of cancer patients. The problem of clinical decision-making pertaining to the management of patients with hematologic cancer is multifaceted and intricate due to the risk of therapy-induced myelosuppression, multiple infections, and febrile neutropenia (FN). Myelosuppression due to treatment increases the risk of sepsis and mortality in hematological cancer patients with febrile neutropenia. A high prevalence of multidrug-resistant organisms is also noted in such patients, which implies that these patients are left with limited or no-treatment options amidst severe health complications. Hence, early screening of patients for such organisms in their bodies is vital to enable hospital preparedness, curtail the spread to other weak patients in hospitals, and limit community outbreaks. Even though predictive models for sepsis and mortality exist, no model has been suggested for the prediction of multidrug-resistant organisms in hematological cancer patients with febrile neutropenia. Hence, for predicting three critical clinical complications, such as sepsis, the presence of multidrug-resistant organisms, and mortality, from the data available from medical records, we used 1166 febrile neutropenia episodes reported in 513 patients. The XGboost algorithm is suggested from 10-fold cross-validation on 6 candidate models. Other highlights are (1) a novel set of easily available features for the prediction of the aforementioned clinical complications and (2) the use of data augmentation methods and model-scoring-based hyperparameter tuning to address the problem of class disproportionality, a common challenge in medical datasets and often the reason behind poor event prediction rate of various predictive models reported so far. The proposed model depicts improved recall and AUC (area under the curve) for sepsis (recall = 98%, AUC = 0.85), multidrug-resistant organism (recall = 96%, AUC = 0.91), and mortality (recall = 86%, AUC = 0.88) prediction. Our results encourage the need to popularize artificial intelligence-based devices to support clinical decision-making.

A review of mathematical model-based scenario analysis and interventions for COVID-19.

Mathematical model-based analysis has proven its potential as a critical tool in the battle against COVID-19 by enabling better understanding of the disease transmission dynamics, deeper analysis of the cost-effectiveness of various scenarios, and more accurate forecast of the trends with and without interventions. However, due to the outpouring of information and disparity between reported mathematical models, there exists a need for a more concise and unified discussion pertaining to the mathematical modeling of COVID-19 to overcome related skepticism. Towards this goal, this paper presents a review of mathematical model-based scenario analysis and interventions for COVID-19 with the main objectives of (1) including a brief overview of the existing reviews on mathematical models, (2) providing an integrated framework to unify models, (3) investigating various mitigation strategies and model parameters that reflect the effect of interventions, (4) discussing different mathematical models used to conduct scenario-based analysis, and (5) surveying active control methods used to combat COVID-19.

Reinforcement learning-based decision support system for COVID-19.

Globally, informed decision on the most effective set of restrictions for the containment of COVID-19 has been the subject of intense debates. There is a significant need for a structured dynamic framework to model and evaluate different intervention scenarios and how they perform under different national characteristics and constraints. This work proposes a novel optimal decision support framework capable of incorporating different interventions to minimize the impact of widely spread respiratory infectious pandemics, including the recent COVID-19, by taking into account the pandemic's characteristics, the healthcare system parameters, and the socio-economic aspects of the community. The theoretical framework underpinning this work involves the use of a reinforcement learning-based agent to derive constrained optimal policies for tuning a closed-loop control model of the disease transmission dynamics.

Crosstalk between HER2 and PD-1/PD-L1 in Breast Cancer: From Clinical Applications to Mathematical Models.

Breast cancer is one of the major causes of mortality in women worldwide. The most aggressive breast cancer subtypes are human epidermal growth factor receptor-positive (HER2+) and triple-negative breast cancers. Therapies targeting HER2 receptors have significantly improved HER2+ breast cancer patient outcomes. However, several recent studies have pointed out the deficiency of existing treatment protocols in combatting disease relapse and improving response rates to treatment. Overriding the inherent actions of the immune system to detect and annihilate cancer via the immune checkpoint pathways is one of the important hallmarks of cancer. Thus, restoration of these pathways by various means of immunomodulation has shown beneficial effects in the management of various types of cancers, including breast. We herein review the recent progress in the management of HER2+ breast cancer via HER2-targeted therapies, and its association with the programmed death receptor-1 (PD-1)/programmed death ligand-1 (PD-L1) axis. In order to link research in the areas of medicine and mathematics and point out specific opportunities for providing efficient theoretical analysis related to HER2+ breast cancer management, we also review mathematical models pertaining to the dynamics of HER2+ breast cancer and immune checkpoint inhibitors.

Optimal adaptive control of drug dosing using integral reinforcement learning.

In this paper, a reinforcement learning (RL)-based optimal adaptive control approach is proposed for the continuous infusion of a sedative drug to maintain a required level of sedation. To illustrate the proposed method, we use the common anesthetic drug propofol used in intensive care units (ICUs). The proposed online integral reinforcement learning (IRL) algorithm is designed to provide optimal drug dosing for a given performance measure that iteratively updates the control solution with respect to the pharmacology of the patient while guaranteeing convergence to the optimal solution. Numerical results are presented using 10 simulated patients that demonstrate the efficacy of the proposed IRL-based controller.

Reinforcement learning-based control of drug dosing for cancer chemotherapy treatment.

The increasing threat of cancer to human life and the improvement in survival rate of this disease due to effective treatment has promoted research in various related fields. This research has shaped clinical trials and emphasized the necessity to properly schedule cancer chemotherapy to ensure effective and safe treatment. Most of the control methodologies proposed for cancer chemotherapy scheduling treatment are model-based. In this paper, a reinforcement learning (RL)-based, model-free method is proposed for the closed-loop control of cancer chemotherapy drug dosing. Specifically, the Q-learning algorithm is used to develop an optimal controller for cancer chemotherapy drug dosing. Numerical examples are presented using simulated patients to illustrate the performance of the proposed RL-based controller.