One Data for Rare Diseases (ODRD) software

Ministry of Health & Family Welfare, Government of India formulated a ‘National Policy for Treatment of Rare Diseases’ (NPTRD) in 2017. This could not even be implemented and yet another new NPTRD was formulated in 2021. That explains the complexities, enigma, dilemmas, and roadblocks of the problem of ‘rare diseases'.

The definition of ‘rare diseases’ is where the problem starts! World Health Organisation (WHO) defines a rare disease when population prevalence is less than 10/10000. But different countries have adopted different levels ranging from 1/10000 (Taiwan) to 6.4/10000 (USA). There is disagreement on whether rare diseases can be defined based solely on prevalence criteria. Other criteria such as severity, life-threatening nature, hereditary relations, treatability, geographic concentration, population genetics, customs, and practices have been proposed by different studies at different times; but no large-scale structured study has been commissioned yet that can support any of these with sufficient data.

Early diagnosis of rare diseases is also a challenge due to a lack of awareness among primary care physicians and a lack of adequate screening and diagnostic facilities. Traditional genetic testing, including Next-generation Sequencing or Chromosomal microarray studies, are too expensive and rarely available to be applied in a large population context. Many doctors lack the appropriate training to be able to correctly and timely diagnose and treat these conditions. It takes patients in the United States (US) an average of 7.6 years and patients in the United Kingdom (UK) an average of 5.6 years to receive an accurate diagnosis, typically involving as many as eight physicians (four primary care and four specialists). In addition, two to three misdiagnoses are typical before arriving at a final diagnosis. 

Given a small patient pool of ‘rare diseases’ there is very little credible data on their pathogenesis, natural history, clinical behaviour, and treatment outcome. The extremely high level of clinical expertise required to treat such diseases and the multi-disciplinary environment are also not available widely in India and most other global south countries. About 95% of rare diseases have no approved treatment, and less than 1 in 10 patients receive disease-specific treatment. Where drugs are available, they are prohibitively expensive (cynically referred to as ‘Orphan drugs’ as hardly any pharma company makes them), and hardly available.

In India, there has been no epidemiological data on the incidence and prevalence of “rare diseases’, so the burden is not even known. There is also the moral dilemma of balancing competing public health and economic interests while spending public money on rare diseases. What will be yielded and then what to do with the patients, given few treatment options; when the country already has much larger public health emergencies such as anemia, undernutrition, etc, where the impact is highly visible, easier to achieve, and reaps demographic dividends.  

It is evident that the problem described above is complex, not well-understood, multi-factorial in origin, and diverse across geographies, ethnic and economic groups. The sector is highly data-thin, and without data, there is no way to address the problem basket.

We now introduce our solution - a deep-learning enabled data-driven approach to the conundrum of ‘Rare diseases’ in India and other resource-constrained geographies in the global south. Based on earlier experience of our team in building similar solutions for COVID and Diabetes Mellitus, we will build ''One Data for Rare Diseases'' (ODRD) software that will enable large-scale screening of target population groups, initially consisting of people in the 0-30 years age group and patients known to be suffering from rare diseases within a defined community. Apart from screening, there will be regular monitoring, earlier detection, risk categorisation, and structured early referral of at-risk/diseased individuals to a network of graded treatment centres. This will be ‘designed with user’, operationally compatible in low band-width conditions, and easy-to-use by community health workers in both rural and urban environments.

The data elements in the software will include demographics of surveyed population, anthropometric measurements, climate conditions, dietary practices, prior use of drugs (prescription/otherwise) and substances, nutritional status, hygiene practices and prior infections, family and past history of rare diseases, current symptoms if any, basic physical examination, etc. Certain relevant and practicable biomarkers will also be tested using frugal point-of-care innovative diagnostic devices for which suitable collaborations have been established with two of the leading technology institutes in India – IIT Kharagpur and IIT Guwahati. 

From the collected data, using statistical and AI-ML models, we can infer characteristics that are associated (including weak association) with the rare disease. For example, if a rare disease is restricted only to a family, then there is a possibility that certain specific single nucleotide polymorphisms (SNPs) may be responsible for this (requires further investigation). We calculate the prevalence of each of the observed rare diseases. Rare disease-specific risk scores will be developed based on the available data. A multivariate analysis, using Mahalanobis distance, will be carried out to find out whether collected dietary, climate, biomarker, proteomic, and clinical data are associated with rare diseases. 

We will also build capacity among community health workers on a pilot scale for software-driven survey work in a target community, undertake basic physical examinations, and conduct diagnostic tests as part of the surveillance program. This work will be monitored by a specialist team of doctors, public health experts, and medical institutions providing clinical care. The data and consequent knowledge will form the backbone of a National policy on Rare Diseases.

AI-ML algorithm driven ODRD software and access to frugal technologies can be a game changer in early detection of rare diseases. For example, ODRD software can better target small geographical areas identified based on prevalence, risk factors and local genetic predispositions in context of inborn errors of metabolism such as congenital hypothyroidism, congenital adrenal hyperplasia and glucose-6-phosphate dehydrogenase deficiency.

On the other side, infectious diseases e.g., Leptospirosis and scrub typhus (for which there is little awareness among primary care doctors) are also expected to be identified at an earlier stage through ODRD software.

Our proposed solution serves multiple stakeholders at multiple levels. 

Patients with rare diseases – Access to treatment is a massive challenge for them, over and above the financial hardship (as treatment of most rare diseases, whenever available, is generally highly complex and expensive). Countries like India can only have few centres of such expertise and complexity. Our complete data-system will connect the right patient to the right treatment facility and provide guidance and access. Follow-up of such patients is also a problem as the local doctor has very little knowledge and expertise. The community health workers, using the software that will have an Electronic Health Record (EHR) component, can support such periodic follow-ups with full documentation and monitor progress.

Community at large – Once adopted, the combination of health workers and the software establishes a mechanism for screening of all target populations in the community. 

Skilled workforce – We have stated in this proposal about the lack of appropriately trained health workforce at the community level who may be tasked with the implementation of screening, earlier detection, and public health programs; and non-specialist doctors (only such doctors are available in peripheral areas in India) being generally unaware of rare diseases and, therefore, failing to detect them early or at all. Through this proposal, we will develop a batch of skilled community health workers on a pilot scale who would be able to undertake all activities (pre-medical intervention stage). The AI-ML engine of the software will provide evidence-based guidance to inexperienced non-specialists to raise suspicion of diagnostic possibility and make suitable graded referrals. The same software will also provide information on linkages of various rare disease problems in different geographies to the nearest referral centre.

National government – We have studied multiple countries in Asia, Western Europe, and North America, and it is encouraging to see that governments are aware of the problem of rare diseases. Some of them (including India) just have no idea about how to proceed with a strategy because of the paucity of information, lack of studies, and a totally data-thin landscape. Our data-driven approach will offer a comprehensive solution to the governments in terms of data on incidence and prevalence, etiological factors and their interrelationships (nutrition, infection, familial, climate, pollution, etc.), stages of disease among the sufferers, their lack of access to medical care, etc. This knowledge will help the government to both formulate and implement appropriate public health policies and set up tertiary centres of excellence for their treatment (India already has a few such hospitals, known as ‘Nidan Kendra’). 

The team members have worked together in an immersive manner on the development of data-science technologies for health systems with a focus on improving access to health for the underserved population. The team lead has extensive experience in developing decision support software engaging the users and rural community, taking bottom up human centred approach which is linguistically familiar, socioculturally congruent. For software development, the ability of the health workers to navigate through the UI, their local language familiarity, health-seeking behaviour of the community, the common symptoms and how people express them all played a pivotal role. The usage of the software is within the capacity of the community health workers and converging with their lives. 

In addition to decision-support systems for clinical care, the team has developed two (2) similar software solutions for Covid Severity Score and Diabetes Risk Score. The core team is supported by a group of specialist doctors from different disciplines with vast experience and leadership positions in their own fields.    

Multi-speciality team – our team draws from a wide cross-section of specialists. They include doctors with vast community medicine experience of working in rural areas, working at the interface between medical science and health technologies. It has engineers and scientists working on innovating deep-science but frugal health technologies, software engineers and data scientists, experts on statistics and mathematics, biologists, social entrepreneurs, and program managers. 

Participation of globally renowned institutions – the team is drawn from some of the ‘Institutes of National Importance’ and ‘Higher Educational Institutions’ in India, such as IIT Guwahati. The mentor group of this project has Prof. Marc Madou of the University of California at Irvine, and Prof. Amitabha Ghosh, Former Director of IIT Kharagpur.

Training model – the team has experience of training a large number of rural youth in different states of India as community health workers. They are certified as per a National Occupation Standard by the Government of India. They are specially trained on digital literacy, technonology usage and financial literacy. The team also has large scale experience in buiding microenterprise, entrepreneurship among the rural youth and driving self-belief among them. 

Technology preparedness – the software team is well-placed to undertake the proposed development involving data analytics, machine learning algorithms, and public health analytics. Omics integration is a future task. Multiple frugal diagnostic devices have been implemented; a pipeline of new innovations is set to roll out as soon as they receive national certification. 

Government support – the project is likely to receive support from various government agencies in India for its alignment with the government policy of 2021. As mentioned earlier, the Government is also looking for a solution. We are confident the Government will adopt this data-driven approach in its national program for the treatment of rare diseases 2021, if we can successfully complete the first phase of development and conduct a pilot-scale implementation model. 

Thousands of health workers – a large number of rural youths, formally trained and certified, may receive employment as community health workers. 

Over the last several years, our team has devoted itself to creating a comprehensive software ecosystem known as 'Uday'. This system is specifically designed to support the operation of eHealth clinics, particularly in rural environments where traditional physician access may be limited. Leveraging community health workers and advanced algorithms, 'Uday' paves the way for remote, efficient, and effective medical care.

At its core, 'Uday' utilizes a sophisticated static medical algorithm that enables Community Health Workers (CHW) to meticulously compile comprehensive patient medical histories. By using a structured set of questionnaires tailored to each 39 commonly reported symptoms, the software guides CHW physical examinations, both general and complaint-based. Upon completing the assessment, 'Uday' synthesizes a Medical History Note (MHN) and Examination History Note (EHN). It further consolidates all gathered data, including demographic details, medical history, family history, and vitals, and forwards it to a backend doctor. After reviewing the data the doctor interacts with patients and the CHWs and writes prescriptions. The software also has separate modules for diagnostic tests, billing etc. LoRaWAN technology is being integrated to serve the areas with no bandwidth. 

The proposed addition of the ODRD module represents a significant enhancement to the 'Uday' ecosystem. This comprehensive software will be deployed for the pilot. This add-on will incorporate modules addressing rare diseases, augmenting the capabilities to cater to an even broader spectrum of health conditions. As part of this initiative, we will establish the prevalence of each observed rare disease and generate risk scores specific to each ailment based on existing data. The existing data generated by the Uday software will also help us understand the target diseases to develop a self-learning algorithm. 

Utilizing advanced analytical techniques, such as multivariate analysis and Mahalanobis distance, we aim to discern any associations between collected dietary, climate, biomarker, proteomic, and clinical data, and these rare diseases. With this multifaceted approach, 'Uday' continually evolves to provide robust, sophisticated solutions for rural communities and thus making our solution a prototype of the aforesaid solution rather than just a concept.

Our tasks, as explained in this proposal, include i) Full development of One Data for Rare Diseases (ODRD) software, including deep-learning algorithms and data analytics, ii) Training an initial batch of community health workers (40), iii) Undertake survey among a target population of 300,000 using ODRD and test its effectiveness.

Based on the outcome of this phase of the work, we will plan a larger multi-centric study targeting a population of about 1-2 million, for which we will approach the national government and other development organisations, CSOs active in such areas. At this stage, global collaborations will be sought from selected institutions in the USA, for collaborative and iterative further knowledge development, best practice sharing, etc. 

We have partially de-risked the software development process through our experience in developing software for Covid Severity Risk Score and Diabetes Risk Score in other projects.

We are really looking at developing a solution for the entire global south, and in stage III, our aim will be to undertake multi-country surveillance study through the ODRD. 

We are applying for the Horizon challenge requesting financial support for software development, and completing the pilot scale survey among 300,000 people through 40 trained health workers. 

We can see our path to establishing the ODRD as a global solution. We are confident if we can complete the first part of this ambitious project, we will receive significant traction from government and development agencies for further progress in Stages II & III.  

Born and raised in a remote village in West Bengal, our Team Lead, Sohom, possesses a profound understanding and personal connection with rural communities. His first-hand experience of the unique challenges these communities face, especially in healthcare, has laid the foundation for the inception and development of our project, 'Uday'.

Over the years, working with NGOs and academia, Sohom has actively worked with these communities, frequently engaging with community members, health workers, doctors, and leaders. This hands-on approach has deepened his grasp of the local needs, preferences, and cultural nuances that significantly influence healthcare practices in these regions. Sohom has made a significant effort to establish and nurture strong networks within these communities. This approach facilitates an ongoing dialogue and feedback loop that allows us to continually refine and adapt 'Uday' to meet the specific needs of the communities we serve. Collaborations with local health officials, non-profit organizations, and community health workers have proven instrumental in implementing our project and making necessary adjustments on the ground.

Beyond this, Sohom has emphasized the importance of ongoing learning and adaptation, integrating his extensive field experiences into our software and hardware development process. This allows us to create intuitive, context-specific solutions that are user-friendly and relevant to the people we aim to serve. As an integral part of both the software and hardware development teams, Sohom's deep understanding of the communities and their healthcare needs has been pivotal in bringing our innovative, point-of-care devices and software solutions to those who need them most.

Sohom’s connection with these communities extends beyond professional boundaries. His commitment to enhancing healthcare access and quality, coupled with his shared history and vision, fosters trust and mutual respect that serves as the cornerstone of our project's success.

The ‘backbone’ of our proposed solution is the software that i) becomes a country-wide repository of comprehensive analytic knowledge of rare diseases in India ii) provides clinical and public health decision-support to medical and public health professionals, and iii) helps build large-scale long-term surveillance programs and public health strategies. There is no such solution available in India today. Given the vast geography, large population, ethnic and linguistic variations, and rare occurrences, any sustainable long-term strategy for addressing rare diseases has to be founded on a data-driven approach in India.

The software’s deep-learning algorithm opens up endless possibilities for global collaboration on biological and behavioural aspects of rare diseases, genetic and phenotypic influences, the role of pre-marital counselling, unknown etiological factors such as diet/frequent infections/hygiene, etc., availability of effective and evidence-based treatment strategies, outcome etc. Such an approach also enhances the opportunities for the introduction of Precision medicine & Gene therapy in the management of rare diseases.

The impact of climate change, if any, on the occurrence of rare diseases has not been studied at all. Our proposed data-driven approach enables the commissioning of such in-depth studies easily and at a low cost.

Easy customizability, agnostic of language/location/band-width, simple UIs, and a ‘design with user’ approach define the software solution that makes deployment across the world simpler. The success of any AI-ML model-based software depends on its easy-to-use features for the end users. In this case, the end user is community health workers. We have already developed (for other projects) Natural Language Processing based algorithm that can convert user inputs (in the vernacular) to appropriate medical terminologies. The AI-ML algorithm is such that it can be used for any language. 

The software can be easily deployed on the field for use by trained health workers. Its analytic engine flags early deviations which may not be clinically detectable at that stage to untrained eyes of non-specialist doctors in the periphery (eg. a subtle anthropometric mismatch in cranial circumference or shape in a newborn indicating a potential rare disease with high morbidity/mortality which would otherwise be missed to naked eye observation) and provides evidence-based medical practice support to the physicians. Earlier detection remains the key for providing optimum patient support and outcome.

It aligns with national government objectives and fills an important void in its armamentarium to tackle rare diseases. Despite enactment of two policies in 2017 and 2021, no measurable progress on the ground has happened yet. 

Our solution brings multiple capacities - survey, risk score, risk category, treatment calibration, structured referral, outcome measurement, long-term follow-up, data-driven decision-support systems under one umbrella - ODRD.

This community health worker-delivered data-driven approach is finally integrated with introduction of frugal diagnostic technologies to the communities that will help earlier detection, monitoring of progress of treatment, outcome and continue to generate valuable health data facilitating global research. The numbers of some of the rare diseases is so small that global data needs to be brought together to make better sense and bring relief to the sufferers.

Year #1:

Full development of ODRD (One Data for Rare Diseases) software with data integration and AI-ML driven analytics.

Training of forty (40) community health workers from among rural youth at Jhargram (Latitude: 22° 27' 13.82" N; Longitude: 86° 59' 41.89" E) & Birbhum (Latitude: 23° 53' 13.614" N, 87° 34' 44.8644"E)

Basis for location selection –

• Extreme poverty
• Vast rural hinterland
• Considerable health gap
• Extreme resource poor settings
• Diverse geography and environment

Demography of the selected districts

Jhargram – Land of forest and rich folk culture; sex ratio - 377; 50% from SCST community, literacy rate 70.92% (female – 61.66%); population density 374/sq km; industrially backward, agriculture is the primary source of earning.

 Birbhum – Land of forest and stone quarry, drought and flood; female and male population is 49% and 51%, respectively; SCST and minority population is 36% and 35%, respectively; literacy rate 71% (female – 64%). Predominantly farmers and agricultural labourers; women artisans are engaged in cottage industry mainly Kantha Stitch and leather handicrafts. 

H. 1 Year #2:

1. Complete baseline survey among a total population of 300,000, covering specifically the target groups (all persons from 0-30 years of age).

2. Basis for population selection: The objective of the first survey will be to i) identify all persons suffering from rare diseases in the surveyed community (vast majority of rare diseases will either manifest within a few years of birth or within young adulthood) ii) cover all young adults in marriageable age groups for genetic/biochemical/anthropometric changes that could affect birth of offsprings with rare diseases and iii) create a database of genetic, habitual, dietary, infections, climate, environment pollution, other diseases etc. for the population covered. This data will be analysed to examine the relationship between increased occurrence of any of these and incidence of rare diseases (retrospective) and prospectively following them up to see how risk scores move up and down.

3. Publish Wave #1 of data analysis report and plan the following – 

a) Linkage with higher centres of treatment so that sufferers are provided with immediate access to care

b) Key indicators for causal factors

c) Key influencers for public health policy formulation

d) Training & awareness creation requirement among health human resources

e) Strategy for ‘Diffusion of innovation’ among community members

H. 2 Year #2:

1 Iterative improvement of the software and identification of diagnostic technology portfolio through stakeholder engagement – the full smart solution (ODRD + Diagnostic + Trained health workers) and present the solution to Government of India

Year #3-#4:

1. Complete similar survey across 5.0 million population

2. Create a cohort of [x] number of patients

3. Continuous follow-up and multiple data collection for analytic conclusions – etiology, climate change & other impacts, disease progress, early identifiers, risk scores, access to treatment, public health policies

Year #4-#5:

1. Adoption by Government of India in its National Program for Treatment of Rare Diseases.

2. Adoption by at least two (2) other countries of significance in South Asia and Africa.  

Goal #1: Full development of ODRD software with complete data integration and analytics, along with AI-ML capabilities.

This software will bring together a multitude of data sources, including clinical records, demographic information, research studies, and patient-reported data, into a comprehensive and easily accessible platform. The software will employ advanced analytics and AI-ML algorithms to identify patterns, trends, and potential therapeutic targets that might otherwise remain hidden. It also facilitates encrypted data sharing, ensuring that knowledge and insights gained from one rare disease can benefit other similar conditions. The software will be able to capture the longitudinal aspect of the rare disease. Therefore, the progression of the rare disease like how fast/slow the disease progress over time can be inferred from the software. 

Goal #2: Training and certification (Government of India) of forty (40) community health workers from among rural youth at Jhargram & Birbhum.

Our collaborating and partner organisation, JSV Innovations Pvt. Ltd., has existing training facilities affiliated with the highest national bodies. We expect the selection of candidates (from among the local communities where survey will be conducted) to be completed within 4 weeks. Classroom, Skill lab training, On-the-job training and Assessment-Certification will be completed within next 6 months (as per National Skill Qualification Framework – NSQF).  

Goal #3: Complete baseline survey among a total population of 300,000, covering specifically the target groups (all persons from 0-30 years of age).

Based on published census date by the government, target population groups will be identified. The total population will be spread over 25-30 Gram Panchayats (last-mile public administration system). Local administrative authorities will be informed and necessary permissions obtained for the survey to commence. This will be conducted in a door-to-door manner by the health workers and the data will be collected into the software (to be developed for this purpose) on a tablet computer on a real-time basis, linking to GPS coordinates. Thus, there will be a continuous monitoring of the process. Each group (2 health workers) will cover 10 households per day (approximately, depending on geographical conditions. 

Goal #4 - Publication of Wave #1 of data analysis, including Cohort formation.

The report will be prepared by the project team consisting of domain experts – doctors, data experts, life-science specialists, statisticians and social work experts. The report will be evaluated by an independent committee that may be constituted for this purpose to ensure its completeness, integrity of data and standard of security, and finally rationale behind conclusions drawn and recommendations made. This will also be the first such cohort in India.

Goal #5: Engage with the national government with the full complement of the solution – functional ODRD software with full heat map and analytic capabilities, Trained team of digitally literate and technology-enabled health workers from local communities, specialist teams and network of treating hospitals (Nidan Kendra), and an established model of access along with community-based monitoring by the same health workers.

The development of ODRD software involves a comprehensive technical approach that encompasses data integration, analytics, and collaboration features. The software utilizes cutting-edge technologies to handle the complexities of rare disease data and provide valuable insights.

The major core technologies that power our solution are: a) Collecting the right data from the target population in a longitudinal manner, b) Secure storage of collected data, c) Individual privacy preserving data sharing for research purposes, d) Use of Statistics, AI-ML to infer about the rare diseases, e) Easy to understand report generation that is free from complexities of the technology used, d) Involving community health workers (who understand the target population best), e) Making the interface of the software easily understandable for health workers using the vernacular inputs.

At its core, the software employs robust data integration techniques to bring together diverse data sources. It leverages standardized data formats and ontologies to harmonize disparate datasets, ensuring compatibility and interoperability. This enables seamless integration of various types of data, such as electronic health records, demographic data, imaging data, and patient-reported outcomes.

To extract meaningful information from the integrated data, ODRD software incorporates advanced analytics capabilities. It utilizes machine learning algorithms, natural language processing, and data mining techniques to identify patterns, correlations, and anomalies within the data. These analytical tools enable researchers and clinicians to uncover hidden relationships, genetic markers, and potential therapeutic targets that can inform the understanding and treatment of rare diseases. The software also focuses on facilitating collaboration among different stakeholders. It provides a user-friendly interface that allows different stakeholders to access and contribute to the platform. It incorporates secure data sharing mechanisms, ensuring privacy and confidentiality while enabling data exchange for research purposes. From a security perspective, ODRD software implements robust security measures (encryption, blockchain, HIPAA etc.) to protect sensitive data. It employs encryption techniques to safeguard data both during transmission and at rest. Access controls, authentication protocols, and audit trails are implemented to ensure authorized access and traceability of data usage.

To enable scalability and performance, the software will be built on a robust infrastructure that will utilize cloud computing resources to handle large volumes of data and accommodate growing user demands. The system architecture will be designed to be flexible and modular, allowing for easy integration of new data sources, target diseases, and functionalities as the field of rare diseases evolves.

Overall, the ODRD software will have robust data integration, analytics, security, and user interface design. By leveraging these technical components, the software enables efficient data management, advanced analytics, and collaboration, ultimately driving progress in rare disease research and improving patient outcomes.

Solution team:

Full-time staff – 4 (Revolut Healthtech Pvt. Ltd.)

Part-time staff – 7 (Doctors & Software development)

Collaborator team – 2 (IIT Guwahati and JSV Innovations Pvt. Ltd.)

Community Health Workers - 40

Key members of the Solution team:

Sohom Banerjee MSc (Applied Physics & Instrumentation Sciences) 

Dr. Bibaswan Basu PhD (Human Physiology), JSV Innovations Pvt. Ltd.

Dr. Sayanti Roy MD (Community Medicine)

Prof. Palash Ghosh (Statistics), IIT Guwahati

Saptaparni Chatterjee BTech (Computer Science)

Dr. Pronil Roy MBBS

Subhojit Pathak BSc, BTech (Software Development) 

Debosmita De MSW, JSV Innovations Pvt. Ltd. 

The team has 5 years of experience of working on clinical decision-support systems including Covid Severity Score (2020-21) and Diabetes Risk Score (2021-22). This particular solution being proposed is a new development and it is at prototype level.

Our solution team consists of eight (8) members with varied age ranging between 23 and 49. Out of the 8 members, three (3) are women and two (2) are in the senior decision-making position. The team members have diverse academic background, diversified expertise with substantial experience in their domain. In spite of the diverse nature of the team, it is the ‘philosophy’ which binds the team and drive it to create a sustainable transformational impact.

The team is driven by the common vision to address the field of ‘Rare Disease’ which is highly complex, poorly understood, data thin, and implement a deep learning algorithm-based data driven approach through ODRD software and make a paradigm shift in the domain of ‘Rare Disease’ from data thin to data driven knowledge repository in context of India.

The community health workers, certified by Govt. of India, mentored by highly experienced doctors and other domain experts in the team, are dedicated, sincere and committed to the vision of the program. Diversity, equity and inclusiveness are three core attributes of the resource pool of community health workers. The trained health workers are from four (4) states of India. These health workers from different states have different culture, language, custom, ethnicity, social adversities,religion etc. Their optimism, esteem, motivation, understanding of their roles and desire for societal contribution have uplifted them as socially productive community. Among our health workers, 65% are women, over 60% are from disenfranchised section of the society. The digitally literate and technology enabled health workers’ familiarity with the local language, understanding about the sociocultural aspects of the community, people health-seeking behaviour and its expression contribute to a great extent in development of the software and designing the holistic 3T model (Training, Technology, Task Shifting).

Through entrepreneurship development, financial literacy and overall capacity building, they form social impact enterprise which enable them to achieve higher social acceptance, financial sustainability and deliver the primary care and public health solutions to the last mile in a seamless manner. This approach enables them to get engaged in the program, without attrition. As they are the daughter and son of the soil, it is easier for them to engage with the community members at large predominantly opinion leaders, village elders and get accepted and trusted by the community as a whole. Primary credibility is one of the core pillars in healthcare.

In a nutshell, the solution team can be characterized as a gender-balanced team with diversity in age, engagement of people from varied expertise, capacity to uphold the entire value chain of ideation to implementation for the last mile. The proposed solution is driven by process, technologies and expert human resources including the trained and certified health workers as the ambassadors represent the diversity, equity and inclusiveness.

Beneficiaries:

i) The local rural population residing in the initial survey area ii) The sufferers from rare diseases identified through the survey iii) Identification of at-risk individuals from within the community and enabling access to  medical advice of health-seeking behaviour change communication iv) health workers, deriving their livelihood through this service model. Indirect beneficiaries include – i) Doctors, access to patients from a wider geography without requiring physical movement ii) Science & Technology researchers for technology innovation and iii) Public health professionals, for policy and implementation.

Partners & Key stakeholders

i) Government – development and implementation of a practicable national policy on management of rare diseases ii) Patients suffering from rare diseases – access to medical care and regular follow-up through treatment institutions; social support through community health workers iii) Local government & administration – alignment for buy-in iv) Centres of excellences – to build/upgrade such centres across different regions of the country to serve as the nodal hospital for treatment of rare diseases identified from the community v) Doctors – willing to be supported by data and evidence-driven management protocols vi) Supervision team – regular technical support, upskilling, community engagement planning vii) Health workers – it is important to motivate them and gain their trust and confidence in this healthcare delivery model.

Key activities

ODRD software will be developed, along with risk score and risk categorisation. 40 young unemployed youth (female>male) will be chosen from the community and trained and certified as community health workers. The health workers will undertake a screening survey among the target population group (all persons 0-30 yrs of age and all known sufferers of rare disease in the community) using the software through door-to-door visits. The data from the software will be analysed by statisticians, doctors and biologists to identify trends, causal factors, ascribe risk score and category. The sufferers will be referred to treatment to higher medical centres (Nidan Kendra); the at-risk sub-group will be subjected to detailed assessment using laboratory and other tests. The sum of ‘at-risk’ groups and known sufferers will be followed up longitudinally as a ‘cohort’ for further studies into diseases causation, how addressing risk factors help movement of patients/at-risk population to a lower risk category. The same technology and process may then be used for a much larger study and gradually absorbed by the government in its national program. 

Channels

i) Local government administration ii) Engaging with influencers such as school teachers, village elders etc. iii) Community-facing activities – focused group discussions, small group public meetings attended by doctors and social workers iv) Involvement of ‘informal practitioners’ by offering them a formal platform for seeking advice on their patients from qualified doctors.

Cost structure:

Capital cost (One-time) will be required for purchasing IT & medical equipment for developing the software, conducting the survey and documenting data. Regular costs will include salaries and wages of IT team, health workers and M&E team. Other costs include travel and accommodation, monitoring & supervision, internet, power, telephone and routine office running expenses.

What we are proposing through this project is to develop a data-driven preventive, curative and promotive approach to ‘Rare diseases’ management - improved understanding, enhanced knowledge, creating local community health human resources, access to treatment, awareness among the peripheral doctors, easier follow-up through risk scoring and categorisation, dynamic heat maps and AI-ML based suggestive clinical plans to doctors. Another key aspect of our proposal is to identify ‘at-risk’ individual young men and women in the community in the marriageable ages and based on software algorithm risk assessment, avoid the risk of bearing newborn with ‘rare diseases’, thus preventing occurrence.

Given our earlier experience of working with similar data systems (Covid, Diabetes, Oral cancer etc,) our team is uniquely poised to be successful in developing the software and the process/network that integrates health workers, community members, doctors and treatment centres. We already have collaborations with at least two IITs in India (Kharagpur and Guwahati) for development of diagnostic technologies that will further enable and strengthen the process. 

The methodology proposed by us can be developed at a fraction of a cost when compared with other public health programs. This is primarily because of its data-driven nature and dependence on frugal diagnostic technologies which will be developed at IITs at their own cost. This proposal will create permanent capacity in the community through trained health workers who may also be used for delivery of other forms of primary care in the community. It establishes a seamless referral system to higher centres which already exist in different parts of the country at no additional cost such as All India Institute of Medical Sciences New Delhi. 

The sustainability of this proposal is fundamentally derived from the fact that the model, (Language-agnostic, Band-width agnostic, Smart clinical algorithm-driven ODRD Software with device integration and data analytics) when adopted by national governments and scaled across the nation, will provide a bulwark of a structure and system that will generate in-depth knowledge about rare diseases, enable monitoring through dynamic heat maps, prevent or lessen occurrence of new disease through appropriate counselling of young adults planning to marry, empower local doctors without specialist knowledge through its AI-ML based inferences and evidence-based suggestions and provide an organised route to referral for higher treatment and information. 

In summary, this proposal is about strengthening national governments in addressing a challenge and reducing burden of diseases and cost of management which oherwise extracts a debilitating and dehumanising cost to the sufferer, family members and immediate society. 

Revolut has raised Venture Capital funding for software development and upgradation of our existing sofware 'Uday' during 2022-23 (USD - $ 31,250) from Pontaq.‘Uday’ was developed in collaboration with John Hopkins University, USA. Financial support was received from Jadavpur University and University of Calcutta towards development of the software. For software upgradation further support has been received from Indian Council of Medical Research (ICMR), Govt. of India during 2021-2022. Regarding implementation of pilot scale digital primary care and public health program, grant was received from Department of Science & Technology, Govt. of India. Our partner organization JSV Innovations Pvt. Ltd. received a mandate from West Bengal Scheduled Castes, Scheduled Tribes and Other Backward Classes Development & Finance Corporation, Govt. of West Bengal during 2018-2019 to train 1246 scheduled caste and scheduled tribe rural youth and employ them in an entrepreneurial model to deliver rural digital primary care and public health. The same ‘Corporation’ has awarded us with a mandate to establish eight (8) digital clinic clusters across five (5) districts of West Bengal during 2019-20 and 2020-21. We have also supported JSV Innovations in executing the mandate received from IIT Kharagpur, IIT Guwahati (Revolut is incubated at IIT Guwahati), Dechi Health Trust, Nigeria etc.