It has been found that taking a look at standard ECG tests, AI can help identify patients who are more likely to die of any medical reason within a year. The researchers from Geisinger Health System in Pennsylvania reached this conclusion after analyzing the results of 1.77 million ECGs and other records from almost 4,00,000 patients. The team of researchers used that data to compare ML-based models. The model can either directly analyze the raw ECG signals or depend on aggregated human-derived measures and commonly diagnose disease patterns. The neural network model was found to be more efficient which can directly analyze the ECG signals for predicting the one-year risk of death.

After looking at standard ECG tests, Artificial Intelligence (AI) can help identify patients most likely to die of any medical cause within a year, claim researchers. To reach this conclusion, researchers from Geisinger Health System in Pennsylvania analysed the results of 1.77 million ECGs and other records from almost 400,000 patients. The team used this data to compare machine learning-based models that either directly analysed the raw ECG signals or relied on aggregated human-derived measures (standard ECG features typically recorded by a cardiologist) and commonly diagnosed disease patterns. The neural network model that directly analysed the ECG signals was found to be superior for predicting one-year risk of death. Surprisingly, the neural network was able to accurately predict risk of death even in patients deemed by a physician to have a normal ECG.

New York– After looking at standard ECG tests, Artificial Intelligence (AI) can help identify patients most likely to die of any medical cause within a year, claim researchers. To reach this conclusion, researchers from Geisinger Health System in Pennsylvania analyzed the results of 1.77 million ECGs and other records from almost 400,000 patients. The team used this data to compare machine learning-based models that either directly analyzed the raw ECG signals or relied on aggregated human-derived measures (standard ECG features typically recorded by a cardiologist) and commonly diagnosed disease patterns. The neural network model that directly analyzed the ECG signals was found to be superior for predicting one-year risk of death. Surprisingly, the neural network was able to accurately predict risk of death even in patients deemed by a physician to have a normal ECG.

New York, After looking at standard ECG tests, Artificial Intelligence (AI) can help identify patients most likely to die of any medical cause within a year, claim researchers. To reach this conclusion, researchers from Geisinger Health System in Pennsylvania analyzed the results of 1.77 million ECGs and other records from almost 400,000 patients. The team used this data to compare machine learning-based models that either directly analyzed the raw ECG signals or relied on aggregated human-derived measures (standard ECG features typically recorded by a cardiologist) and commonly diagnosed disease patterns. The neural network model that directly analyzed the ECG signals was found to be superior for predicting one-year risk of death. Surprisingly, the neural network was able to accurately predict risk of death even in patients deemed by a physician to have a normal ECG.

After looking at standard ECG tests, Artificial Intelligence (AI) can help identify patients most likely to die of any medical cause within a year, claim researchers. To reach this conclusion, researchers from Geisinger Health System in Pennsylvania analyzed the results of 1.77 million ECGs and other records from almost 400,000 patients. The team used this data to compare machine learning-based models that either directly analyzed the raw ECG signals or relied on aggregated human-derived measures (standard ECG features typically recorded by a cardiologist) and commonly diagnosed disease patterns. The neural network model that directly analyzed the ECG signals was found to be superior for predicting one-year risk of death. Surprisingly, the neural network was able to accurately predict risk of death even in patients deemed by a physician to have a normal ECG.

May 11, 2017 – Studying human diseases is the equivalent of solving a massive and dynamic jigsaw puzzle with pieces that are constantly changing shape. A team involving researchers from the Nutritional Immunology and Molecular Medicine Laboratory (NIMML) at Virginia Tech and the Biomedical and Translational Informatics (BTI) Institute at Geisinger Health System are working together to advance precision medicine by integrating clinical data, artificial intelligence (AI) systems, and advanced machine-learning (ML) methods. In a new study, the collaborative team of experts have developed new computational methods to stratify stroke patients in an emergency setting, paving the way to data-driven triage process with higher fidelity. The rich, longitudinal data warehouse of the Geisinger Health System (GHS) has detailed electronic health records (EHR) of over 3 million active participants. This rich data is one of the major strengths that allowed Geisinger to be selected to participate in the national Precision Medicine Initiative (PMI) Cohort Program with the goal of improving the ability to prevent and treat diseases based on individual differences in lifestyle, environment, and genetics.

This work is motivated by the needs of predictive analytics on healthcare data as represented by Electronic Medical Records. Such data is invariably problematic: noisy, with missing entries, with imbalance in classes of interests, leading to serious bias in predictive modeling. Since standard data mining methods often produce poor performance measures, we argue for development of specialized techniques of data-preprocessing and classification. In this paper, we propose a new method to simultaneously classify large datasets and reduce the effects of missing values. It is based on a multilevel framework of the cost-sensitive SVM and the expected maximization imputation method for missing values, which relies on iterated regression analyses. We compare classification results of multilevel SVM-based algorithms on public benchmark datasets with imbalanced classes and missing values as well as real data in health applications, and show that our multilevel SVM-based method produces fast, and more accurate and robust classification results.