PreSco – Neonatal Sepsis Platform

Delayed diagnosis and low access to specialized healthcare in rural areas are some of the chief causes for high neonatal mortality in India. More than 40% global neonatal deaths are reported to occur within the first 24 hours of birth in India primarily due to low resources and time lost in transportation to urban centres where most healthcare specialists are located.

There is a strong felt need for rapid and innovative screening systems which can address the existing gaps in the early diagnosis space. Use of new-age artificial intelligence and machine learning technologies can bridge this gap in early screening of neonatal infections. However, there is a significant skill among frontline healthcare workers in usage of Mobile health and telemedicine applications. Our Solution focuses on equipping necessary digital skills in ASHA (accredited social health activists) Frontline healthcare workers in India by using PreSCo, our AI-ML based Neonatal Sepsis Prediction application. 

Every year, more than 6 lakhs new born babies die in India of which 30% are due to neonatal sepsis. India has a high neonatal mortality rate of 25. India also has the global highest first 24-hour neonatal deaths. More than 40% neonatal deaths are reported to occur within the first 24 hours of birth in developing countries like India chiefly due to low resources and time lost in transportation to urban centers. Under staffing in primary health centers is an additional concern as 8% of them do not having clinical staff and 39% have no lab technicians. Doctor to patient ratio in the country is one of the lowest at 1:10000 against the WHO’s stipulation of 1:1000. To solve the problem of acute lack of health resources in rural India, ASHA (Accredited Social Health Activists) volunteers act as the first point of contact between communities and healthcare systems and over the years have become the frontline access points for public health in India. There are close to nine lakh women ASHA workers in India. Geographically, they are best placed to serve as the first-level human interface between doctors and technology on one end and the patients and communities on the other. However, they lack the necessary digital or technical skills to operate digital health applications and required to be reskilled. Our solution will address the problem of neonatal mortality through implementation of digital health solutions such as ours through intensive training of frontline health workers.

The main beneficiary groups of the solution are:

• Frontline Health Workers such as ASHA Workers in India and associated Rural Clinics
• Mother & Children: Young mothers and infants who are in need of quality healthcare touch

Accredited Social Health Activists (ASHAs) volunteers are the first point of contact between communities and the Indian healthcare system. They are women who are usually residents of the villages that they serve in. Over the years since their introduction in 2005, they have emerged as the focal point of public health program implementation across India. ASHA volunteers are responsible for referral service for reproductive and child health, nutrition, health education, and promoting universal immunisation, amongst others. However, they are largely underutilised despite training as there are gaps in their knowledge and skills particularly related to digital health applications, telemedicine etc.

 Our digital health application and associated digital reskilling program will enable these frontline community health workers to cater to issues related to early identification and reporting of neonatal sepsis infections and health complications. These skills should also help them fulfil higher-order roles like telemedicine coordinators in low technology and low resource settings. As digital health coordinators, they will also be equipped with positive behavioural skills to help them have empathetic and impactful interactions with patients. Through our skills upgradation program, Frontline healthcare workers will be empowered to be autonomous for on-ground decision-making and handling exceptions in the space of maternal and neonatal health.

Our solution is aligned to the challenge as it addresses issues related to knowledge and skill gaps among frontline health workers and equips them with the necessary technical and behavioural skills to handle advanced digital health applications in low resource settings. The solution will be initially piloted in India and can be scaled up to implement in countries like Indonesia.

Frontline Health workers will be trained on our digital health application, PreSco, which is an integrated cloud-based AI-ML platform using evidence-based medicine and has the capability of addressing existing issues of low resources, accessibility and affordability, inability to provide timely diagnosis, medical errors and difficulty in manually processing large amount of data in real time in healthcare.

We have designed a comprehensive training and reskilling program for ASHA Frontline Health workers in India to use the PreSco application for data entry, uploading, processing of data, interpretation of data and transmitting data to a tertiary care centre or Neonatal Specialist. The training protocol includes run through of the application, features, medical terms used in the application, FAQs on operating the application, data entry and final submission for score risk generation and subsequent referral from specialists. The training protocol also includes required behavioural skills in interfacing with patients and doctors.

We in the process of tying up with various state governments in India for testing and validation of PreSco application and are targeting to train at least 50 – 100 ASHA Frontline health workers in each region in the next 8 – 10 months.

Dr Hyma Goparaju is CEO & Co-founder of Avyantra Health Technologies. She is an industry professional with close to 20 years of experience and is a PhD from IBS, Hyderabad. She is also Program Leader the company’s graduation from Unicef Innovation Fund’s Data Science global cohort of 2019.

An integrated cloud-based Digital health platform that works on machine learning technologies is capable of addressing existing issues of low resources, accessibility and affordability, inability to provide timely diagnosis, medical errors and difficulty in manually processing large amount of data in real time in healthcare. The wide range of input parameters used for predicting risk score enable multiple types of algorithms (non-invasive and invasive) depending on the resource availability at the health centre (ranging from frontline health volunteer to a tertiary health centre). Our digital platform facilitates rapid assessment, early screening and triaging of babies. Cloud platform and open-source technologies enable accessibility, interoperability and affordability. The platform can be accessed by a wide range of devices such as mobiles, desktops etc by frontline health volunteers who are the first level health providers in the maternal and infant space. Advanced machine learning algorithms are capable of handling large amount of longitudinal data (clinical symptoms, blood test reports, Images etc) and can unfold complex interactions between multiple variables. Predictive algorithms can predict the onset of infections like neonatal sepsis, pneumonia etc much before the clinical manifestation of symptoms. Deep learning algorithms can speed up confirmatory diagnosis based on images such as X-rays, CT scans etc due to their faster training ability with fewer rules and aid in administration of effective and timely treatment. Advanced digital platforms can address shortage of medical specialists and radiologists in various regions and serve as effective medium for delivering equitable health services.

Predictive Algorithms Levels – Frontline Healthcare workers, Primary Health Centre (PHC) & Tertiary Health Service levels

Score generation time    < 2 minutes

Risk Assessment Levels on a scale of 0 to 10

Low (0-5) – Green colour coded

Medium (5.1 – 8) – Orange colour coded

High (8.1 – 10) – Red colour coded

Technology Stack        

Database – MySQL;

Frontend – HTML5, Bootstrap 4, CSS 3, Angular 7;

Cloud Platform – AWS EC2, API – Node JS;

Coding – Python; Machine Learning Algorithms – Decision Trees, Random Forest, XG Boost

Accuracy Achieved       70-80%

Platform Link    http://presco.avyantra.com/#/

Digital health can aid in improved and cost-effective decision making. The socio-economic impact of our solution is long-lasting and sustainable. Experiments from developed countries have demonstrated that use of digital health platforms have the potential of reducing infections by 84% and antibiotic usage by 50%, impacting not just health outcomes but also reducing healthcare costs. In India, ASHA workers (Accredited Social Health Activist) are community health workers instituted by the Ministry of Health and Family Welfare as a part of the National Rural Health Mission. They are frontline health volunteers who are also entrusted with the responsibility of early maternal and child healthcare. However, the tools they are equipped with are minimal, lacking the capacity to handle healthcare challenges especially those brought in by post COVID-19 pandemic such as lockdowns, social distancing etc. At many places, there are about 9 ASHA workers for every 10,000 people increasing the burden on them. Digital applications and tools can ease the existing constraints and drastically improve quality of health delivery, generate equity in an unbalanced healthcare system, especially in developing or low-resource countries. 

Digital health market is expected to go up to $150 billion by 2030 and save at least $50 billion in healthcare expenses in India alone.

NEAT TERM -

In India, a Frontline health volunteer has access to at least 1000 to 10,000 mothers and babies in a year which varies according to the region they cover. Going by the minimum and maximum coverage, with training offered to about 50 to 100 ASHA volunteers, our platform can impact 50,000 to 100,00,00 mothers and babies. We plan to achieve these goals in the near term.

LONG TERM -

SCALE THROUGH LARGE RURAL DEMAND – Every year, more than 25 million babies are born in India of which 10 million are located at the bottom of the pyramid. Rural market in India is very large with more than 60% of the population lives in rural areas. The large size of the neonatal market (USD 30 to 35 billion) lends sustainability to our solution.

SCALE THROUGH OPERATIONAL AND FINANCIAL GROWTH BY PARTNERSHIPS - . We are in final rounds of a couple of Funding Initiatives by the Govt of India for Startups and are looking to improve our reach to various markets. We have a LONG-TERM AGREEMENT (LTA) with Unicef for supplying our solution to their member countries.

SCALE THROUGH COST EFFECTIVENESS - Demand for quality healthcare services in rural India is extremely high. The large demand will help us sustain our model and scale it.

We have well defined processes in place for monitoring and evaluation such as setting Project Goals and Objectives, Defining Indicators, Data Collection Methods, Data Analysis and Reporting, Dissemination of Information through communication protocols agreed.

Data Quality Assurance Mechanisms – We follow data quality assurance plan with defined process for ensuring data validity, integrity, reliability, precision and timeliness.

Frequency of Appraisals – Daily basis for project operations, monthly for comprehensive project review and quarterly for review of project goals and impact.

Results Framework – We follow a well-defined template for result assessment. The results framework includes a comprehensive performance management plan for monitoring of key project goals and objectives as agreed with our partners.

Measurable Indicators -

We are targeting to train atleast 50 - 100 ASHA Frontline health workers in each of the state government rural health centers for which are in the process of formal tie-ups. 

Percent Achievement of Key Indicators for Sepsis Score - We target to achieve 80 - 90% for key metrics such as Accuracy, Sensitivity, Specificity for risk scoring algorithms. Number of referrals, infection rates, antibiotics used, diagnosis to treatment turnaround time etc used etc are among others.

Financial Barrier – We are in need of consistent funding support to enhance the features of our platform, and complete testing and validation for launch and commercialisation.

Technical Barrier - Data availability in required digital formats for retrospective study as well as conversion of unstructured medical data to structured data for analysis is a complex task.

Legal & Policy Barrier - Data privacy laws are stringent across the globe with lack of uniformity at regional levels.

Infrastructure Barrier - Lack of required internet bandwidth in developing countries and less developed countries is an issue in the short term.

Resources Available - We are pursuing our project goals with support from international agencies like UNICEF and government support programmes for start-ups. However, we are in need of consistent funding partnerships to create sustainable impact. 

Financial Barrier – We are working with several funding agencies, both domestic as well as international ones, to meet our future funding requirements.

Technical Barrier - We are working to overcome this challenge with advanced technical architecture for big data, image analysis and data mining techniques.

Legal & Policy Barrier - We are taking steps to adhere to data privacy laws to ensure safety and privacy of patient data.

Infrastructure Barrier – Though we face low internet connectivity issues, however, since telecom infrastructure is improving in many developing countries we expect it will support usage of digital products through mobile applications.

Resources Available - We are pitching our solution at various forums to improve access to resources and increase our partnership scope.  

Our team consists of 5 members - 2 full time and 3 part time. 

Dr Hyma Goparaju, CEO & Co-Founder, Avyantra Health Technologies (Full Time)

KVKLN Rao, COO, Founder, Avyantra Health Technologies (Full Time) 

Prashanth Krishnan – Technical Architect 

Dr M Alimelu, Head of Department, Niloufer Hospital, Hyderabad (Advisor)

Dr U Dinesh Kumar, Professor of Statistics & Head DCAL, Indian Institute of Management, Bangalore (Advisor) 

Avyantra Health Technologies was founded in 2017 and is based at Hyderabad, India. The co-founder Hyma Goparaju is an industry professional with close to 20 years of experience and is a PhD from IBS, Hyderabad. The second founder KVKLN Rao is also an industry professional with over 20 years of experience and is currently a healthcare entrepreneur who is also pursuing part-time PhD in Data Science from IIM Shillong to further his passion for Data Science and its application in healthcare. They are advised by experienced domain experts from healthcare and data analytics. Avyantra’s investment partners are well-established national and international institutions. The team has a healthy balance and members are committed to the goals.

Currently, our project team has a good balance of established industry partners fully committed to the cause of the project. A healthy balance of male and female members across the partner organizations at all levels, right from roles requiring key decision-making to execution skills has been maintained. We have plans to recruit specialists directly as well as on a project basis to meet our short- and long-term objectives. For the same, a manpower of about 15 to 20 staff is planned for.

So far, we have worked with two large hospitals for development of our algorithms. Going forward, we propose to test our platform and algorithms in multiple centres. For this, we have recently tied up with two state governments for piloting our solution in low resource centres. We hope to execute these pilots in 2022 and achieve generalization of our algorithms and test and validate our digital platform.

Our business model is centred around three types of algorithms that use both non-invasive clinical parameters and invasive parameters such as blood tests. Our platform provides affordable and accessible evidence-based medicine to enable good outcomes from digital health. Our predictive algorithms aid in early and effective triaging of babies. An affordable EMR platform for both hospitals as well as parents is yet another novel offering of PreSco. The value propositions of our solution are 1. An integrated EMR and Risk scoring platform with three different levels of algorithms that work as per resource availability since not all healthcare centres in our country are equipped with similar level of diagnostic facilities, and 2. An affordable EMR platform for hospitals and patients that is accessible from mobiles.

We are looking for a greater visibility for our solution. We believe that collaboration with MIT Solve will help us in gaining the same. Additionally, sharing and transfer of experience, cross collaborations and funding access would help us to scale our platform better.

We are in the process of tying up with state governments in India for piloting and testing. Our start-up is one of the three winners of the BIRAC NASSCOM Jancare Challenge 2021, a nation-wide challenge for health startups in India. As a part of the award, we have won an opportunity to pitch our solution to state governments across India. So far, we have received positive confirmation from two states for piloting our solution. We require more such partnerships for our platform. We also require funding support for piloting in these places. These pilots will help us progress towards commercialization. Additionally, we need to comply with NDHM’s (national digital health mission) guidelines to achieve regulatory requirements for our platform. We look forward to achieving all of these with partnerships and funding support.