Modeling physician variability to prioritize relevant medical record information

The Learning Electronic Health Record

Electronic medical records (EMRs) constitute the electronic version of all medical information included in a patient's paper chart. The electronic health record (EHR) technology has witnessed massive expansion in developed countries and to a lesser extent in underresourced countries during the last 2 decades. We will review factors leading to this expansion, how the emergence of EHRs is affecting several health-care stakeholders; some of the growing pains associated with EHRs with a particular emphasis on the delivery of care to the critically ill; and ongoing developments on the path to improve the quality of research, health-care delivery, and stakeholder satisfaction.

Maternal Preconception Body Mass Index Overtakes Age as a Risk Factor for Gestational Diabetes Mellitus

Introduction-The purpose of this study was to determine the relative impact of modifiable and non-modifiable risk factors in the development of gestational diabetes mellitus (GDM), with a particular focus on maternal preconception body mass index (BMI) and age, two important determinants of insulin resistance. Understanding the factors that contribute most to the current escalation of GDM rates in pregnant women could help to inform prevention and intervention strategies, particularly in areas where this female endocrine disorder has an elevated prevalence. Methods-A retrospective, contemporary, large population of singleton pregnant women from southern Italy who underwent 75 g OGTT for GDM screening was enrolled at the Endocrinology Unit, "Pugliese Ciaccio" Hospital, Catanzaro. Relevant clinical data were collected, and the characteristics of women diagnosed with GDM or with normal glucose tolerance were compared. The effect estimates of maternal preconception BMI and age as risk factors for GDM development were calculated through correlation and logistic regression analysis by adjusting for potential confounders. Results-Out of the 3856 women enrolled, 885 (23.0%) were diagnosed with GDM as per IADPSG criteria. Advanced maternal age (≥35 years), gravidity, reproductive history of spontaneous abortion(s), previous GDM, and thyroid and thrombophilic diseases, all emerged as non-modifiable risk factors of GDM, whereas preconception overweight or obesity was the sole potentially modifiable risk factor among those investigated. Maternal preconception BMI, but not age, had a moderate positive association with fasting glucose levels at the time of 75 g OGTT (Pearson coefficient: 0.245, p < 0.001). Abnormalities in fasting glucose drove the majority (60%) of the GDM diagnoses in this study. Maternal preconception obesity almost tripled the risk of developing GDM, but even being overweight resulted in a more pronounced increased risk of developing GDM than advanced maternal age (adjusted OR for preconception overweight: 1.63, 95% CI 1.320-2.019; adjusted OR for advanced maternal age: 1.45, 95% CI 1.184-1.776). Conclusions-Excess body weight prior to conception leads to more detrimental metabolic effects than advanced maternal age in pregnant women with GDM. Thus, in areas in which GDM is particularly common, such as southern Italy, measures aiming to counteracting maternal preconception overweight and obesity may be efficient in reducing GDM prevalence.

A highly predictive cardiac positron emission tomography (PET) risk score for 90-day and one-year major adverse cardiac events and revascularization

BACKGROUND

With the increase in cardiac PET/CT availability and utilization, the development of a PET/CT-based major adverse cardiovascular events, including death, myocardial infarction (MI), and revascularization (MACE-Revasc) risk assessment score is needed. Here we develop a highly predictive PET/CT-based risk score for 90-day and one-year MACE-Revasc.

METHODS AND RESULTS

11,552 patients had a PET/CT from 2015 to 2017 and were studied for the training and development set. PET/CT from 2018 was used to validate the derived scores (n = 5049). Patients were on average 65 years old, half were male, and a quarter had a prior MI or revascularization. Baseline characteristics and PET/CT results were used to derive the MACE-Revasc risk models, resulting in models with 5 and 8 weighted factors. The PET/CT 90-day MACE-Revasc risk score trended toward outperforming ischemic burden alone [P = .07 with an area under the curve (AUC) 0.85 vs 0.83]. The PET/CT one-year MACE-Revasc score was better than the use of ischemic burden alone (P < .0001, AUC 0.80 vs 0.76). Both PET/CT MACE-Revasc risk scores outperformed risk prediction by cardiologists.

CONCLUSION

The derived PET/CT 90-day and one-year MACE-Revasc risk scores were highly predictive and outperformed ischemic burden and cardiologist assessment. These scores are easy to calculate, lending to straightforward clinical implementation and should be further tested for clinical usefulness.

Application of Machine Learning to Improve Appropriateness of Treatment in an Orthopaedic Setting of Personalized Medicine

The rise of personalized medicine and its remarkable advancements have revealed new requirements for the availability of appropriate medical decision-making models. Computer science is an area that plays an essential role in the field of personalized medicine, where one of the goals is to provide algorithms and tools to extrapolate knowledge and improve the decision-support process. The minimum clinically important difference (MCID) is the smallest change in PROM scores that patients perceive as meaningful. Treatment that does not achieve the minimum level of improvement is considered inappropriate as well as a potential waste of resources. Using the MCID threshold to identify patients who fail to achieve the minimum change in PROM that results in a meaningful outcome may aid in pre-surgical shared decision-making. The decision tree algorithm is a method for extracting valuable information and providing further meaningful information to the domain expert that supports the decision-making. In the present study, different tools based on machine learning were developed. On the one hand, we compared three XGBoost models to predict the non-achievement of the MCID at six months post-operation in the SF-12 physical score. The prediction score threshold was set to 0.75 to provide three decision-making areas on the basis of the high confidence (HC) intervals; the minority class was re-balanced by weighting the positive class to penalize the loss function (XGBoost cost-sensitive), oversampling the minority class (XGBoost with SMOTE), and re-sampling the negative class (XGBoost with undersampling). On the other hand, we modeled the data through a decision tree (assessment tree), based on different complexity levels, to identify the hidden pattern and to provide a new way to understand possible relationships between the gathered features and the several outcomes. The results showed that all the proposed models were effective as binary classifiers, as they showed moderate predictive performance both regarding the minority or positive class (i.e., our targeted patients, those who will not benefit from surgery) and the negative class. The decision tree visualization can be exploited during the patient assessment status to better understand if those patients will benefit or not from the medical intervention. Both of these tools can come in handy for increasing knowledge about the patient’s psychophysical state and for creating an increasingly specialized assessment of the individual patient.

Evaluation of eye tracking for a decision support application

Eye tracking is used widely to investigate attention and cognitive processes while performing tasks in electronic medical record (EMR) systems. We explored a novel application of eye tracking to collect training data for a machine learning-based clinical decision support tool that predicts which patient data are likely to be relevant for a clinical task. Specifically, we investigated in a laboratory setting the accuracy of eye tracking compared to manual annotation for inferring which patient data in the EMR are judged to be relevant by physicians. We evaluated several methods for processing gaze points that were recorded using a low-cost eye-tracking device. Our results show that eye tracking achieves accuracy and precision of 69% and 53%, respectively compared to manual annotation and are promising for machine learning. The methods for processing gaze points and scripts that we developed offer a first step in developing novel uses for eye tracking for clinical decision support.