Detection of pulmonary hypertension using EHR
Machine Learning is reaching the tipping point in many aspects of the medical field, from disease diagnosis to drug development. In that row, there is a recent breakthrough in the machine learning model will help to predict the progressive disease called "pulmonary hypertension" in the early stage using real-world Electronic Health Records(EHR) datasets.
Before jumping into the research aspects, we need to know what exactly pulmonary hypertension is and why we want to predict the disease in the earlier stage.
The challenging factor of Pulmonary Hypertension:-
Pulmonary hypertension (PH) is a type of high blood pressure that damages the arteries in our lungs, and it can lead to heart failure if not diagnosed and treated early. But PH can only be interpreted slowly. It is a challenging way to analyze the PH disease in its early stage.
The symptom of a PH patient is a hurdle to find until the disease progresses. However, the new novel machine learning model developed may help detect PH in the early stages and improve patient outcomes.
Building machine learning model using electronic health records:-
Earlier ML models were used to run a single test or only for a specific PH subgroup. However, the first predictive algorithm to predict PH was performed well in detection of two treatable PH subgroups using the EHR database.
This will ease the patient from going through many tests and help to detect PH symptoms for many patients simultaneously.The predictive ML algorithm has studied around 11 million patients' EHR data between 2007 to 2019.
With the use of a given EHR database, the ML model was able to identify 116,000 patients with PH disease. The time duration would have been more if they had done it in the traditional approach rather than using the algorithm to test the patients.
The ML model could predict PH symptoms up to 18 months before clinical diagnosis.
Early Detection of Symptoms using ML:-
According to the research paper that has been published in the International Journal of Cardiology, the predictive algorithm can detect diastolic heart failure and valve disorders symptoms of PH.
The area under the receiver operating characteristic curve of the predicted PH is 0.92. And AUROCs were 0.94, 0.89, 0.86, and 0.83 for intervals of 2 to 6, 6 to 12, 12 to 18, and for 18 months, approximately.
Using EHR records, it is possible to find that the early signs of a PH patient have become viable and can support the physicians in treating the patient in effect.
How Does ML Elevate EHR?
Technologies like the predictive algorithm are evolving each year and can play a significant role in identifying and diagnosing deadly diseases during their early stage.
Many healthcare industries have undergone massive transformations in recent years by adopting the latest technologies, like artificial intelligence, data science, etc., in their EHRs.
Using Machine learning models, it becomes easy to analyze data, identify patterns, and provide personalized treatment recommendations that can help improve patient outcomes.
Machine learning plays important role like:-
- Extraction of relevant data from patient records and other sources.
- Finding missing values and formatting the data in a way suitable for machine learning algorithms.
- Predict a patient's risk of developing a particular disease, and the relevant features might include age, medical history, and lifestyle factors.
Thus, machine learning in EHR products can provide numerous benefits, better patient outcomes, improved efficiency, and cost savings for healthcare providers and organizations.
Check our EHR product which provides you an easy-to integrate option with better efficiency and cost saving offers. Vozo EHR software has in-built templates which will help your workflow to be done smoothly and save your time.
Customize the data like you wanted and study your patient records in a detailed manner with our friendly user interface software.
