Inspiration
With smart hospitals and healthcare, and as a result of the use of new technologies, the amount of health data from patients is increasing at an exponential rate. Due to the exponential increase in data from healthcare facilities, organizations are rethinking medical data storage strategies. For access to medical records and secure storage, an active archival system is very important in order to reduce maintenance costs for legacy systems.
In few years to come, estimates put the amount of data that will be generated on a daily basis at nearly 500 exabytes of data. Some of the new data will be stored short term, a lot of the data will need to be stored long-term which is quantified to be over 500% increase from the past few years.
A single patient within the healthcare ecosystem and industry generate over 70 megabytes of data each year. The generated data includes electronic medical record, voice and imaging data. The amount of patient health data is increasing exponentially, which means the amount of legacy EHR data is skyrocketing as well.
Data collection in healthcare allows health systems to create holistic views of patients, personalize treatments, advance treatment methods, improve communication between doctors and patients, and enhance health outcomes. Analytics can help healthcare organizations harness "big data" to create actionable insights, and improve outcomes.
What it does
EdgeHealth solution consists of different software modules such as a radiology module for hospital staff members to upload imaging files for patients. One prime example is the use in radiography and storage of COVID-19 chest x-rays. By combining this type of data with machine learning, detection of COVID-19 in an X-ray can be automated using Keras, TensorFlow, and deep learning. Another important module in EdgeHealth is the use conversational AI to help patients on the hospital beds and make life easier.
How we built it
I combined various technologies like CORTX object storage software, S3, postgre database, artificial intelligence, Node/express, Vue, React, Flask, Python to build the solution.
Challenges we ran into
One of the major challenges I ran into was the use of the Node.JS SDK for making a connection to CORTX. Making a connection was very easy to use with the python SDK otherwise known as boto3
Accomplishments that we're proud of
Deeper understanding of object storage and development of healthcare solution that can improve the lives of patients and hospital staff members.
What we learned
I did a lot of research around smart hospitals and healthcare as well as new and emerging technologies. I learnt a lot about CORTX and the object storage technology.
Working with the CORTX increased my understanding of distributed object storage system and the numerous APIs associated with an S3-compatible object storage software.
What's next for EdgeHealth
Due to HIPAA regulations, it will be nice to create a solution that allows conversational dialogs within the hospital facility without sending data to the public cloud by bring AI services used in the public cloud to edge. The conversational AI currently is receiving text messages. I want to make it hands-free and only have a patient or doctor converse with voice. In addition, I want to collect COVID-19 patient x-ray dataset with the software and use the data to classify x-ray results. I'm also planning to extend the lessons learnt to the hospitality and automobile industries.
Built With
- cortx
- dialogflow
- flask
- google-dialogflow
- javascript
- node/express
- postgresql
- python
- s3
- vue
Log in or sign up for Devpost to join the conversation.