Swasthya Mohalla (Healthy Neighborhoods)

radiology dataset anonymized and results compared with ASDI OpenAQ
Bulk query and retrieve operation of radiology datasets
EHR record via Orthanc which can be hosted on AWS Ec2
Anonymized patient record which can be hosted on AWS Ec2
bulk query and retrieve transaction service
ethereum scan page
ethereum transaction

Swasthya Mohalla Clinic (Healthy Neighborhoods)

Enabling the bottom of pyramid through EHR (Electronic Health Records) and Radiology services on the cloud using Amazon Comprehend, ASDI UCSC genome browser sequence and annotations (https://github.com/awslabs/open-data-registry/blob/main/datasets/ucsc-genome-browser.yaml), ASDI OpenAQ ((https://github.com/awslabs/open-data-registry/blob/main/datasets/openaq.yaml), Octave/Matlab, OpenTok, Genome Link API, Near, ICON and Ethereum blockchain network and open source PACS and spreadsheet. EHR as a service aims at providing secure data storage, transparent data movement and data authenticity. . Ability to plan food and nutrition using OpenTok and GenomeLink API, access UCSC genome browser sequence and annotations (https://github.com/awslabs/open-data-registry/blob/main/datasets/ucsc-genome-browser.yaml), using EHR's medical report, medical data for better healthcare delivery can be devised and implemented.

AWS Comprehend: We are using Amazon Comprehend Medical to detect and return medical information from the following AWS apps developed by us:

Medication Log: http://aspiringapps.com/web/home/medication-log.html
Medical Notes: http://aspiringapps.com/web/home/medical-notes.html
Medical Invoice: http://aspiringapps.com/web/home/medical-invoice.html
Patient Sign In: http://aspiringapps.com/web/home/patient-signin.html
Diabetic Plus: http://aspiringapps.com/web/home/diabetic-plus.html
Blood Sugar Log: http://aspiringapps.com/web/home/blood-sugar.html
Nutrition Tracker: http://aspiringapps.com/web/home/nutrition-tracker.html

Unstructured clinical text such as physicians notes, discharge summaries, test results, and case notes can be derived using AWS Comprehend. Amazon Comprehend Medical uses natural language processing (NLP) models to detect entities, which are textual references to medical information such as medical conditions, medications, or Protected Health Information (PHI). Please visit: https://github.com/seetadev/EtherGlance/tree/main/amazon-comprehend-medical-document-processing.

Inspiration

EHR (Electronic Health Records) Need in India: The current status of the healthcare sector in India is associated with low public spending (1% of GDP), high out of pocket payments (71%), a high level of anemia among young women (56%), high infant mortality (47/1,000 live births), high maternal mortality (212/100,000 live births).

This highlights the need for major reforms in the healthcare sector. Though information and communication technology (ICT) cannot work as a panacea, it can provide a cost-effective solution to improve the quality and coverage.

Indian doctors are in the process of adopting electric health/medical records maintenance online. It not only gives complete information about the patient, about his previous medical history, and also saves time while providing treatment, as going through the medical record doctor can get in to fast conclusion on what treatment to provide immediately. Indian government is taking consideration into maintenance of health records online.

The proliferation of mobile phones and the availability of high-speed Internet offer the possibilities to provide healthcare services in rural, remote areas of the country.

What it does

Swasthya Mohalla is a community technology cloud solution designed specifically to address the needs of the medical eco-system comprising of hospitals, medical laboratories, patients, diagnostic labs, R&D organizations and improve the operational efficiencies.

Just in Time service Availability of patient’s medical records, genome data across different stakeholder through secure blockchain network and AWS cloud.

Record Management Quality documentation reduces the issues regarding prescription of incorrect medicines. Ability to plan food and nutrition using AWS Comprehend, GenomeLink API, EHR's medical report, UCSC genome browser sequence and annotations (https://github.com/awslabs/open-data-registry/blob/main/datasets/ucsc-genome-browser.yaml), statistics on air quality mapped with chest X Rays using ASDI OpenAQ (https://github.com/awslabs/open-data-registry/blob/main/datasets/openaq.yaml) and Orthanc PACS system, medical data for better healthcare delivery can be devised and implemented.

Research Research laboratories can use the data for computer-aided diagnosis of diseases, suggesting personalized care to patients. Further, food and nutrition methods could be researched and devised for speedy recovery of the patient and prevention of diseases in patients' life after successful treatment.

Web Application Cloud-based web application with OpenTok APis, a material design application linking to GenomeLink API, AWS Comprehend, UCSC genome browser sequence and annotations (https://github.com/awslabs/open-data-registry/blob/main/datasets/ucsc-genome-browser.yaml), ASDI OpenAQ (https://github.com/awslabs/open-data-registry/blob/main/datasets/openaq.yaml) and for quick medical attention.

Transparency Insurance agencies can utilize the data to provide customized health insurance plans to the customer. Personal dietary consultants can provide better food and nutrition plans for the speedy recovery of patients.

Better Cure Journal of medical records covering complete patient history and Genome data using GenomeLink API and weather conditions mapped to chest X Rays using ASDI OpenAQ (https://github.com/awslabs/open-data-registry/blob/main/datasets/openaq.yaml) to improve the cure outcomes.

How we built it

We engineered and developed cutting edge technology on ethereum and Near blockchain network using NuCypher protocol, AWS Comprehend APIs, ASDI UCSC genome browser sequence and annotations (https://github.com/awslabs/open-data-registry/blob/main/datasets/ucsc-genome-browser.yaml), ASDI OpenAQ (https://github.com/awslabs/open-data-registry/blob/main/datasets/openaq.yaml), Octave/Matlab, OpenTok and GenomeLink API, open source spreadsheet and open source PACS (Picture Archiving and Communication Systems) solution:

Creation of radiology files in DICOM format, examination reports and patient’s medical records from hospital visits. DICOM stands for Digital Imaging and Communications in Medicine. This file format is created to distribute and view medical images using a standardized file format. We utilize GenomeLink API and integrate with EHRS platform for better understanding the genetic make-up of a person and suggest appropriate treatment.

Cleansing and validation of in-bound patient data available as medical images and examination reports Authentication, authorization using unique SSN (social security number) for legitimately storing the medical records.

EHR movement across different hospitals, laboratories, pharmacies on requirement basis using:

i) ORTHANC as an open source and light weight DICOM server for medical imaging, querying and retrieving patient’s data and radiology reports from a medical institution.

ii) An inbound web application using AWS Sagemaker, Amazon S3, Amazon Athena, AWS forecast, Amazon API gateway, Application server and OpenTok APIs for viewing medical examination reports and patient records.

AWS Comprehend: We are using Amazon Comprehend Medical to detect and return medical information from the following AWS apps developed by us:

Medication Log: http://aspiringapps.com/web/home/medication-log.html
Medical Notes: http://aspiringapps.com/web/home/medical-notes.html
Medical Invoice: http://aspiringapps.com/web/home/medical-invoice.html
Patient Sign In: http://aspiringapps.com/web/home/patient-signin.html
Diabetic Plus: http://aspiringapps.com/web/home/diabetic-plus.html
Blood Sugar Log: http://aspiringapps.com/web/home/blood-sugar.html
Nutrition Tracker: http://aspiringapps.com/web/home/nutrition-tracker.html

Using Ethereum blockchain:

We are utilizing a variety of key Ethereum tools and libraries: metamask, truffle, embark, nvm, node.js, filecoin, polygon, NFT.Storage, NFTPort, Sequence, 0vix, Chainlink, Fluence, Spheron, Tableland, ipfs and ICON services and cross chain tooling. Our solution offers an Artificial Intelligence-based object detection system that utilizes blockchain solutions for sorting information obtained from a variety of data management bodies, PACS systems deployed at monitoring spots.
Expert.ai NL APIs: We are using Expert.ai NL APIs for getting insights from radiology reports: https://github.com/seetadev/EtherGlance/tree/main/expert-ai-nlp

Using Near and ICON blockchain:

Near protocol integration: Fluence JS peer based on the NEAR API JS and distributed Wasm services wrapping the NEAR RPC API. We are utilizing Near blockchain Protocol to receive feedback on medication log reports by medical experts, store feedback details from both experts and experienced doctors in order to enable early stage detection and prevention of diseases.
We are also utilizing Near Protocol to receive feedback reports by medical experts, store feedback details from both experts, doctors on the reports in order to enable early stage prevention of diseases.
ICON cross-chain services for transferring crypto between ICON and other blockchains like BSC: https://github.com/seetadev/EtherGlance/tree/main/icp/cross-chain-asset-transfer

Purpose

Our solution, EHR as a service, can help in:

Legitimate storage of medical records, invoices using open source PACS and IPFS for storage and react based spreadsheet application.

Records movement in a secure way using NuCypher Re-encryption Protocol, OpenTok APIs, Query and Retrieve application on bootstrap and Orthanc DICOM open source solution.

Preventive treatment and remediation by studying the Genome using the GenomeLink API and planning essential nutrients in the patients' meal, food and nutrition plan.

Challenges we ran into

AWS deployment: We were using PV (paravirtual) based classic instance. We were required to convert PV (paravirtual) instance to an HVM (Hardware Virtual Machine) and convert it from a classic instance to a VPC instance. We faced a number of challenges in converting our PV instance to an HVM instance as our instance was not reachable via ssh. We arrived at a good conclusion on the issue by doing the following:

Created an ami of current classic instance.

Launch a new instance from this ami after shifting to vpc as the base with new VPC security groups and updated volume.

The newly launched instance is working and we are able to ssh and has the code base. We purchased the developer support plan using AWS credits and have being shared key pointers on converting our VPC based PV instance to a VPC based HVM instance. This challenge would not have been solved without the support of AWS team.

The third party administrators (TPAs) have to manually upload the medical history data to create the medi-claim/insurance based NFT. It is not a problem for a small set of vehicles where we can use csv to json conversion using open source libraries like phpexcel in ethercalc but problematic for larger number of patients. At this juncture, the TPAs have to manually connect their wallet and create the medi-claim/insurance claim NFT.

We wish to mention that integrating and sending medical reports using NuCypher protocol required lot of mentorship from Bogdan Opanchuk and Dr. Michael from NuCypher. They helped us at every step of implementing NuCypher whether it was installing the correct versions of requisite libraries and tools and helping us running the re-encryption protocol. We wish to thank the EthIndia community and fellow participant hackers, who helped us learn and implement our idea in a fine way.

Enablement

EHRs can help in data cleansing, creating EHR, records legibility and coordination with hospitals, pharmacies, diagnostic clinics, in case when movement of records is needed. The EHR application can be accessible using Internet browser, all the data is on the cloud. Motivational benefits of our solution include -

Patient information can be secured and shared anywhere at any time with help of EHR. Staff and clinicians can make the direct entry, reduces transcription cost. Quality documentation can be maintained (perceptible, readable, records, reports and charts). Reduce the issues of prescription of incorrect medicines. Ability to plan food and nutrition using GenomeLink API, EHR's medical report, medical data for better healthcare delivery can be devised and implemented. Complete patient history records improve the cure outcomes. These systems greatly enables the physician for immediate patient treatment by capturing the key records. EHR records are the key data for implementing the clinical decision support system.

Business Benefits

a. Patients

Make instant appointments with a doctor and get quick access to e-records. Computer-aided detection will help in early prediction and diagnosis of diseases. Ability to plan food and nutrition using GenomeLink API, EHR's medical report, medical data for better healthcare delivery can be devised and implemented. b. Doctors and Hospitals

Prioritize and handle appointments better. Issue prescriptions Coordinate with specialists. Access to accurate records. Improved care. Better dietary feedback based on genomelink API c. Administrator and Insurers

Connect all stakeholders Personalize care treatment. Accurate and timely payments Reduce the cost of centralized systems d. Employees

Administer Benefit. Reduce healthcare costs. Reward employees Offer affordable benefit packages. Better diet and food programs at offices. e. Research Institutions

Computer-aided detection of diseases and development of prevention models. Personalized medicine and drug discovery practice.

A greater and more seamless flow of information within a digital drug discovery infrastructure, created by electronic medicine trial and test records as a service (EMTTRs), encompasses and leverages digital progress and can transform the way medicines are developed, tested and distributed to improve the global health economy and achievement of Sustainable Development Goals in Healthcare. Our blockchain solution, EHR as a service, can help in:

Legitimate storage of medical and environment records, medical invoices and environment data using ICON services, open source PACS and IPFS for storage and react based spreadsheet application. Records movement in a secure way using NuCypher Re-encryption Protocol, Query and Retrieve application on bootstrap and Orthanc DICOM open source solution. Health counseling and environment awareness steps, preventive treatment and remediation over the ethereum blockchain network using a decentralized twitter application and Embark tools. EHRs can help in data cleansing, creating EHR, records legibility and coordination with hospitals, schools, pharmacies, diagnostic clinics, in case when movement of records is needed. The EHR application can be accessible using Internet browser, all the data is on the cloud.

What's next for Swasthya Mohalla

We are evolving the solution in the following phases:

PHASE I – Requirements Analysis and Design (1 months)

Demonstrate the solution using spreadsheet and PACS software on cloud connected devices, Android phones, iPhones, first generation tablets.

Procuring server hosting for storing and utilizing images and associated video data to prove that real time monitoring is viable.

Set up timeline for planning the 2022 winter deployments and user training session

PHASE II (winter deployments and user training session) (8 months) -Create PoC based on designed specification. Complete the design of web interface.

Initial user testing, and POC refinement -Prototype Release

Start the pilot trials with 2-3 vendors. -Prototype Manufacturing at the Vendors location.

-Prototype validation and assembly: Final Prototype assembly and validation; Refinements based on the mechanical prototype; Final Engineering CAD release.

Manage and provide the hands-on task of exporting images and video reports from the customer to SEETA medical cloud system.

Create and deploy a gateway service in the customer geo-location that will enable the continuing export of images and video reports to the SEETA medical cloud system.

Completion of supporting collateral required to fulfil services and deliverables such as the equipment, supplies and other open source software tools.

Continue the collection, data organization and management of images and associated video report data to improve computer aided detection using deep learning algorithms and integrate them with the platform.

Survey on community’s needs, user interaction, selection of vendors and quality diagnostic centers where we could deploy full-scale pilot, possibly focused on mobile-platform organize a hardware agnostic program

Enable pilot users to be developers of web based platform and contribute in improving the existing deep learning algorithms using websites like Kaggle.

Focus on making the platform interoperable with a variety of vendor systems in different housing societies.

PHASE III (6 months)

Winter 2022-2023 deployment in hospitals, diagnostic labs, clinics, pharma companies and drug design and development institutions.

Our solution is useful only to the extent it is used by the medical eco-system and healthcare community. Thus, we are working with health specialists around the world to focus on these learning challenges:

To make our project freely and readily available to learners everywhere To explore and share best practices To provide a forum for discussion and support for technology for learning To provide mechanism for evaluation and dissemination of results.