Inspiration

According to the study by NHLBI 963+ million, Number of Patients Suffering from Sleep Apnea around the World African Americans are 88%more prone to Sleep Apnea as compared to White Americans 95% of African-American Patients are undiagnosed

As per recent studies, the African American population is at the highest risk of OSA due to the anatomy and overweight Excessive sleepiness, severe snoring, choking while sleeping is the main symptoms of Sleep Apnea. Due to delayed diagnosis, can lead to chronic and fatal health conditions like Diabetes, Stroke. It can be an indicator of Alzheimer's, as per a new study. The pain points include difficulty in early diagnosis leading to surgical complications. Hectic procedures at the Sleep Labs and most importantly Delayed diagnostic, due to RACIAL DISCRIMINATION

What it does

Early diagnostic App for Obstructive Sleep Apnea detection that also helps us to monitor the the severity of the disease

It’s pretty handy to use. The patient clicks on the app, answer a few simple questions, and on the basis of that you get a report whether or not you have sleep apnea.

We will be using Machine Learning (SVM) algorithm for finding out the AHI(The apnea-hypopnea index), a scale that tells whether or not you have Sleep Apnea and how severe it is.

How we built it

We apply a modern machine learning method, the support vector machine, and TensorFlow (open-source software library) to establish a predicting model for the severity of OSA.

we establish a prediction model for Asians by taking body shape profiles and age into account via SVM. We hypothesize that the established predictors are accurate for OSA severity. To confirm this hypothesis, we collected two large patient databases from two independent sleep labs and designed a prediction model for the OSA severity based on body shape profiles.

We randomly partitioned the data into the training dataset and the testing dataset, with the training dataset being created by randomly selecting 20% of the patients while the remainder serving as the testing dataset.

The suspected patient logs in to an app. He’s required to answer questions of BQ like Snoring behavior, obesity, etc. Followed by Epworth Sleep Scale and finally Anthropometric measures like Neck Size, Buttock Size, BMI. On the basis of this information, ML algorithms will calculate AHI by taking reference from the data of the diseased African Americans procured at the backend from the Sleep labs. The final result will be a predictive analysis.

Accomplishments that we're proud of

Comparative market analysis shows OSASense is cost-effective, Makes patients self-dependent. It is the only SCREENING TOOL available in the market for Sleep Apnea and uses Anthropometric features

What's next for OSASense

To scale it to more hence preventing Racial Discrimination. Further research is also needed to identify biomarkers for patients at risk for consequences of OSA as well as predictors for treatment success to enable the development of individual therapeutic approaches for the implementation of personalized medicine.

Share this project:

Updates