A machine learning framework supporting prospective clinical decisions applied to risk prediction in oncology

Predicting surgical outcomes of acute diffuse peritonitis: Updated risk models based on real-world clinical data

Aim

The existing predictive risk models for the surgical outcome of acute diffused peritonitis (ADP) need renovation by adding relevant variables such as ADP's definition or causative etiology to pursue outstanding data collection reflecting the real world. We aimed to revise the risk models predicting mortality and morbidities of ADP using the latest Japanese Nationwide Clinical Database (NCD) variable set.

Methods

Clinical dataset of ADP patients who underwent surgery, and registered in the NCD between 2016 and 2019, were used to develop a risk model for surgical outcomes. The primary outcome was perioperative mortality.

Results

After data cleanup, 45 379 surgical cases for ADP were derived for analysis. The perioperative and 30-day mortality were 10.6% and 7.2%, respectively. The prediction models have been created for the mortality and 10 morbidities associated with the mortality. The top five relevant predictors for perioperative mortality were age >80, advanced cancer with multiple metastases, platelet count of <50 000/mL, serum albumin of <2.0 g/dL, and unknown ADP site. The C-indices of perioperative and 30-day mortality were 0.859 and 0.857, respectively. The predicted value calculated with the risk models for mortality was highly fitted with the actual probability from the lower to the higher risk groups.

Conclusions

Risk models for postoperative mortality and morbidities with good predictive performance and reliability were revised and validated using the recent real-world clinical dataset. These models help to predict ADP surgical outcomes accurately and are available for clinical settings.

Adherence to oral anticancer treatments: network and sentiment analysis exploring perceived internal and external determinants in patients with metastatic breast cancer

PURPOSE

Adherence to oral anticancer treatments (OATs) is a critical issue in metastatic breast cancer (MBC) to enhance survivorship and quality of life. The study is aimed to analyze the main themes and attributes related to OATs in MBC patients. This research is part of a project titled "Enhancing Therapy Adherence Among Metastatic Breast Cancer Patients" designed to produce a predictive model of non-adherence, a decision support system, and guidelines to improve adherence to OATs.

METHODS

The study consists of an exploratory observational and qualitative analysis using a focus group method. A semi-structured interview guide was developed to handle relevant OAT themes. Wordcloud plots, network analysis, and sentiment analysis were performed.

RESULTS

Nineteen female MBC patients participated in the protocol (age mean 55.95, SD = 6.87). Four main themes emerged: (theme 1) individual clinical pathway; (theme 2) barriers to adherence; (theme 3) resources to adherence; (theme 4) patients' perception of new technologies. The Wordcloud and network analysis highlighted the important role of treatment side effects and the relationship with the clinician in the modulation of adherence behavior. This result is consistent with the sentiment analysis underscoring patients experience fear of issues related to clinical values and ineffective communication and discontinuity of the doctor in charge of the patient care.

CONCLUSION

The study highlighted the key role of the individual, relational variables, and side effects as internal and external determinants influencing adherence to MBC. Finally, the opportunity offered by eHealth technology to connect with other patients with similar conditions and share experiences could be a relief for MBC patients.

Predicting Blood Pressure After Nitroglycerin Infusion Dose Titration in Critical Care Units: A Multicenter Retrospective Study

Critical care nurses use physiological indicators, such as blood pressure, to guide their decision-making regarding the titration of nitroglycerin infusions. A retrospective study was conducted to determine the accuracy of systolic blood pressure predictions during nitroglycerin infusions. Data were extracted from the publicly accessible eICU program database. The accuracy of a linear model, least absolute shrinkage and selection operator, ridge regression, and a stacked ensemble model trained using the AutoGluon-Tabular framework were investigated. A persistence model, where the future value in a time series is predicted as equal to its preceding value, was used as the baseline comparison for model accuracy. Internal-external validation was used to examine if heterogeneity among hospitals could contribute to model performance. The sample consisted of 827 patients and 2541 nitroglycerin dose titrations with corresponding systolic blood pressure measurements. The root-mean-square error on the test set for the stacked ensemble model developed using the AutoGluon-Tabular framework was the lowest of all models at 15.3 mm Hg, equating to a 22% improvement against the baseline. Internal-external validation revealed consistent accuracy across hospitals. Further studies are needed to determine the impact of using systolic blood pressure predictions to inform nurses' clinical decision-making regarding nitroglycerin infusion titration in critical care.

Just how transformative will AI/ML be for immuno-oncology?

Immuno-oncology involves the study of approaches which harness the patient's immune system to fight malignancies. Immuno-oncology, as with every other biomedical and clinical research field as well as clinical operations, is in the midst of technological revolutions, which vastly increase the amount of available data. Recent advances in artificial intelligence and machine learning (AI/ML) have received much attention in terms of their potential to harness available data to improve insights and outcomes in many areas including immuno-oncology. In this review, we discuss important aspects to consider when evaluating the potential impact of AI/ML applications in the clinic. We highlight four clinical/biomedical challenges relevant to immuno-oncology and how they may be able to be addressed by the latest advancements in AI/ML. These challenges include (1) efficiency in clinical workflows, (2) curation of high-quality image data, (3) finding, extracting and synthesizing text knowledge as well as addressing, and (4) small cohort size in immunotherapeutic evaluation cohorts. Finally, we outline how advancements in reinforcement and federated learning, as well as the development of best practices for ethical and unbiased data generation, are likely to drive future innovations.

Approach to machine learning for extraction of real-world data variables from electronic health records

Background: As artificial intelligence (AI) continues to advance with breakthroughs in natural language processing (NLP) and machine learning (ML), such as the development of models like OpenAI's ChatGPT, new opportunities are emerging for efficient curation of electronic health records (EHR) into real-world data (RWD) for evidence generation in oncology. Our objective is to describe the research and development of industry methods to promote transparency and explainability. Methods: We applied NLP with ML techniques to train, validate, and test the extraction of information from unstructured documents (e.g., clinician notes, radiology reports, lab reports, etc.) to output a set of structured variables required for RWD analysis. This research used a nationwide electronic health record (EHR)-derived database. Models were selected based on performance. Variables curated with an approach using ML extraction are those where the value is determined solely based on an ML model (i.e. not confirmed by abstraction), which identifies key information from visit notes and documents. These models do not predict future events or infer missing information. Results: We developed an approach using NLP and ML for extraction of clinically meaningful information from unstructured EHR documents and found high performance of output variables compared with variables curated by manually abstracted data. These extraction methods resulted in research-ready variables including initial cancer diagnosis with date, advanced/metastatic diagnosis with date, disease stage, histology, smoking status, surgery status with date, biomarker test results with dates, and oral treatments with dates. Conclusion: NLP and ML enable the extraction of retrospective clinical data in EHR with speed and scalability to help researchers learn from the experience of every person with cancer.

Predicting Acute Care Events Among Patients Initiating Chemotherapy: A Practice-Based Validation and Adaptation of the PROACCT Model

PURPOSE

Acute care events (ACEs), comprising emergency department visits and hospitalizations, are a priority area for reduction in oncology. Prognostic models are a compelling strategy to identify high-risk patients and target preventive services, but have yet to be broadly implemented, partly because of challenges with electronic health record (EHR) integration. To facilitate EHR integration, we adapted and validated the previously published PRediction Of Acute Care use during Cancer Treatment (PROACCT) model to identify patients at highest risk for ACEs after systemic anticancer treatment.

METHODS

A retrospective cohort of adults with a cancer diagnosis starting systemic therapy at a single center between July and November 2021 was divided into development (70%) and validation (30%) sets. Clinical and demographic variables were extracted, limited to those in structured format in the EHR, including cancer diagnosis, age, drug category, and ACE in prior year. Three logistic regression models of increasing complexity were developed to predict risk of ACEs.

RESULTS

Five thousand one hundred fifty-three patients were evaluated (3,603 development and 1,550 validation). Several factors were predictive of ACEs: age (in decades), receipt of cytotoxic chemotherapy or immunotherapy, thoracic, GI or hematologic malignancy, and ACE in the prior year. We defined high-risk as the top 10% of risk scores; this population had 33.6% ACE rate compared with 8.3% for the remaining 90% in the low-risk group. The simplest Adapted PROACCT model had a C-statistic of 0.79, sensitivity of 0.28, and specificity of 0.93.

CONCLUSION

We present three models designed for EHR integration that effectively identify oncology patients at highest risk for ACE after initiation of systemic anticancer treatment. By limiting predictors to structured data fields and including all cancer types, these models offer broad applicability for cancer care organizations and may offer a safety net to identify and target resources to this high risk.

Risk Factors for Return to the Emergency Department and Readmission in Patients With Hospital-Diagnosed Advanced Lung Cancer

BACKGROUND

Advanced lung cancer (ALC) is a symptomatic disease often diagnosed in the context of hospitalization. The index hospitalization may be a window of opportunity to improve care delivery.

OBJECTIVES

We examined the patterns of care and risk factors for subsequent acute care utilization among patients with hospital-diagnosed ALC.

RESEARCH DESIGN, SUBJECTS, AND MEASURES

In Surveillance, Epidemiology, and End Results-Medicare, we identified patients with incident ALC (stage IIIB-IV small cell or non-small cell) from 2007 to 2013 and an index hospitalization within 7 days of diagnosis. We used a time-to-event model with multivariable regression to identify risk factors for 30-day acute care utilization (emergency department use or readmission).

RESULTS

More than half of incident ALC patients were hospitalized around the time of diagnosis. Among 25,627 patients with hospital-diagnosed ALC who survived to discharge, only 37% ever received systemic cancer treatment. Within 6 months, 53% had been readmitted, 50% had enrolled in hospice, and 70% had died. The 30-day acute care utilization was 38%.Small cell histology, greater comorbidity, precancer acute care use, length of index stay >8 days, and prescription of a wheelchair were associated with higher risk of 30-day acute care utilization. Age >85 years, female sex, residence in South or West regions, palliative care consultation, and discharge to hospice or a facility were associated with lower risk.

CONCLUSIONS

Many patients with hospital-diagnosed ALC experience an early return to the hospital and most die within 6 months. These patients may benefit from increased access to palliative and other supportive care during index hospitalization to prevent subsequent health care utilization.