FHIR-PYrate: a data science friendly Python package to query FHIR servers

Use of Real-World FHIR Data Combined with Context-Sensitive Decision Modeling to Guide Sentinel Biopsy in Melanoma

Background: To support clinical decision-making at the point of care, the "best next step" based on Standard Operating Procedures (SOPs) and actual accurate patient data must be provided. To do this, textual SOPs have to be transformed into operable clinical algorithms and linked to the data of the patient being treated. For this linkage, we need to know exactly which data are needed by clinicians at a certain decision point and whether these data are available. These data might be identical to the data used within the SOP or might integrate a broader view. To address these concerns, we examined if the data used by the SOP is also complete from the point of view of physicians for contextual decision-making. Methods: We selected a cohort of 67 patients with stage III melanoma who had undergone adjuvant treatment and mainly had an indication for a sentinel biopsy. First, we performed a step-by-step simulation of the patient treatment along our clinical algorithm, which is based on a hospital-specific SOP, to validate the algorithm with the given Fast Healthcare Interoperability Resources (FHIR)-based data of our cohort. Second, we presented three different decision situations within our algorithm to 10 dermatooncologists, focusing on the concrete patient data used at this decision point. The results were conducted, analyzed, and compared with those of the pure algorithmic simulation. Results: The treatment paths of patients with melanoma could be retrospectively simulated along the clinical algorithm using data from the patients' electronic health records. The subsequent evaluation by dermatooncologists showed that the data used at the three decision points had a completeness between 84.6% and 100.0% compared with the data used by the SOP. At one decision point, data on "patient age (at primary diagnosis)" and "date of first diagnosis" were missing. Conclusions: The data needed for our decision points are available in the FHIR-based dataset. Furthermore, the data used at decision points by the SOP and hence the clinical algorithm are nearly complete compared with the data required by physicians in clinical practice. This is an important precondition for further research focusing on presenting decision points within a treatment process integrated with the patient data needed.

Addressing the Contrast Media Recognition Challenge: A Fully Automated Machine Learning Approach for Predicting Contrast Phases in CT Imaging

OBJECTIVES

Accurately acquiring and assigning different contrast-enhanced phases in computed tomography (CT) is relevant for clinicians and for artificial intelligence orchestration to select the most appropriate series for analysis. However, this information is commonly extracted from the CT metadata, which is often wrong. This study aimed at developing an automatic pipeline for classifying intravenous (IV) contrast phases and additionally for identifying contrast media in the gastrointestinal tract (GIT).

MATERIALS AND METHODS

This retrospective study used 1200 CT scans collected at the investigating institution between January 4, 2016 and September 12, 2022, and 240 CT scans from multiple centers from The Cancer Imaging Archive for external validation. The open-source segmentation algorithm TotalSegmentator was used to identify regions of interest (pulmonary artery, aorta, stomach, portal/splenic vein, liver, portal vein/hepatic veins, inferior vena cava, duodenum, small bowel, colon, left/right kidney, urinary bladder), and machine learning classifiers were trained with 5-fold cross-validation to classify IV contrast phases (noncontrast, pulmonary arterial, arterial, venous, and urographic) and GIT contrast enhancement. The performance of the ensembles was evaluated using the receiver operating characteristic area under the curve (AUC) and 95% confidence intervals (CIs).

RESULTS

For the IV phase classification task, the following AUC scores were obtained for the internal test set: 99.59% [95% CI, 99.58-99.63] for the noncontrast phase, 99.50% [95% CI, 99.49-99.52] for the pulmonary-arterial phase, 99.13% [95% CI, 99.10-99.15] for the arterial phase, 99.8% [95% CI, 99.79-99.81] for the venous phase, and 99.7% [95% CI, 99.68-99.7] for the urographic phase. For the external dataset, a mean AUC of 97.33% [95% CI, 97.27-97.35] and 97.38% [95% CI, 97.34-97.41] was achieved for all contrast phases for the first and second annotators, respectively. Contrast media in the GIT could be identified with an AUC of 99.90% [95% CI, 99.89-99.9] in the internal dataset, whereas in the external dataset, an AUC of 99.73% [95% CI, 99.71-99.73] and 99.31% [95% CI, 99.27-99.33] was achieved with the first and second annotator, respectively.

CONCLUSIONS

The integration of open-source segmentation networks and classifiers effectively classified contrast phases and identified GIT contrast enhancement using anatomical landmarks.

Optimizing platelet transfusion through a personalized deep learning risk assessment system for demand management

ABSTRACT

Platelet demand management (PDM) is a resource-consuming task for physicians and transfusion managers of large hospitals. Inpatient numbers and institutional standards play significant roles in PDM. However, reliance on these factors alone commonly results in platelet shortages. Using data from multiple sources, we developed, validated, tested, and implemented a patient-specific approach to support PDM that uses a deep learning-based risk score to forecast platelet transfusions for each hospitalized patient in the next 24 hours. The models were developed using retrospective electronic health record data of 34 809 patients treated between 2017 and 2022. Static and time-dependent features included demographics, diagnoses, procedures, blood counts, past transfusions, hematotoxic medications, and hospitalization duration. Using an expanding window approach, we created a training and live-prediction pipeline with a 30-day input and 24-hour forecast. Hyperparameter tuning determined the best validation area under the precision-recall curve (AUC-PR) score for long short-term memory deep learning models, which were then tested on independent data sets from the same hospital. The model tailored for hematology and oncology patients exhibited the best performance (AUC-PR, 0.84; area under the receiver operating characteristic curve [ROC-AUC], 0.98), followed by a multispecialty model covering all other patients (AUC-PR, 0.73). The model specific to cardiothoracic surgery had the lowest performance (AUC-PR, 0.42), likely because of unexpected intrasurgery bleedings. To our knowledge, this is the first deep learning-based platelet transfusion predictor enabling individualized 24-hour risk assessments at high AUC-PR. Implemented as a decision-support system, deep-learning forecasts might improve patient care by detecting platelet demand earlier and preventing critical transfusion shortages.