Fetosense

Every year, ~78,000 still births occur in Bangladesh and most neonatal deaths occur due to prematurity. These conditions are linked to high-risk pregnancies. Yet, regular fetal heart rate (FHR) monitoring is absent in most pregnancies, because cardiotocography (the standard used in hospitals) has poor sensitivity; it is capital-intensive, not portable and requires expertise.

Our solution, ‘Fetosense’ (within a mobile program, CareMother) is an AI-based, wireless, portable, smartphone integrated, and affordable fetal monitor. We are developing auto-interpretation of FHR, and monitoring of uterine contractions to predict premature labour. Frontline workers can administer Fetosense, which has been designed for their easy usage, and record data in a smartphone to transfer results in real-time to specialist doctors for critical decisions.

There are 5 million annual pregnancies in Bangladesh and at least 20% are high-risk. Due to the shortage of gynaecologists, Fetosense can enable frontline workers to monitor fetal health and reduce mortality.

Every year, there are 0.5 million still births and 3.5 million premature births in India (our current area of work), while in Bangladesh, ~78,000 babies are still born and ~0.6 million are premature (leading to deaths and long-term developmental problems). These disorders are linked to antenatal care. Studies show that regular antenatal care by bridging gaps through technology, reporting results in-real time to doctors, ensuring follow-ups and effective labour monitoring can reduce mortality. About 99% of these deaths occur in rural and low-income urban settings. However, in Bangladesh, there are only 1646 obstetricians and gynaecologists (registered in the federation), who work mainly in urban hospitals, and only 5.5 doctors and 2.4 nurses per 10,000 people. Moreover, only 31% of women receive optimum antenatal care due to the workload on health workers, logistical difficulties, and absence of effective tools to monitor risks. Cardiotocography or ‘Toco’, used in hospitals to detect labour contractions and monitor fetal heart rate, has a sensitivity of only 68% to assess contractions and 6% to predict premature birth. In addition, Toco is expensive, bulky, needs substantial expertise and high maintenance, and not portable. Thus, mainly the higher income urban women avail effective antenatal risk monitoring.

Our ‘CareMother’ program includes doorstep antenatal tests by frontline workers using portable kits and a smartphone application to share real-time data with doctors. Since 2015, we have registered 30,000+ pregnancies (12000+ high-risk) via 260 health workers across ten states of India, in rural and urban areas. We will pilot Fetosense within CareMother in two urban facilities (for slum population) and two rural sites serving 4153 pregnancies (39.6% high-risk) and having ~100 indirect beneficiaries i.e. health workers, midwives and doctors.

Our needs assessment with 25 urban gynaecologists and 22 rural health centres found a definite scope for an automated digital solution due to the cost, space constraints, and long set-up time for conventional cardiotocography with insignificant return-of-investment, and the absence of portable, scalable fetal monitoring due to a shortage of gynaecologists. Therefore, Fetosense, if used by physicians and health workers, will build key referral linkages with remote specialists.

Twenty urban facilities are using Fetosense since its launch in June 2019 via a revenue-sharing per-test based model to avoid capital investment by the doctor. Fetosense adds to the revenue through digital and convenient fetal distress tests, and doctors are able to share a portion of that revenue over time with the company.

‘Fetosense’ is a non-invasive, wireless, portable, smartphone integrated and affordable AI based fetal heart rate (FHR) monitor. We are developing an auto-interpretation system of FHR records (using Dawes-Redman criteria) to measure fetal cardiac function while reducing the dependence on the specialist doctor, and a tool to measure uterine electrical activity to predict premature labour. Based on incoming data, we are developing a prediction algorithm to assess the consequences of poor FHR on other critical physiological parameters of the fetus. Fetosense has three distinct strengths:

1. It provides an end-to-end digital automated point-of-care solution, which can be used in low-income settings for monitoring fetal health in high-risk and routine pregnancies.

2. Simple usage - health workers and midwives will only need to place a belt on the abdomen and start the device, without repositioning, and receive flagged fetal cardiac and associated risks, which they can report.

3. Due to its portability, low maintenance and linkage with a smartphone, Fetosense can be used for remote monitoring by specialists, and in busy clinics for multiple patients.

Fetosense can be deployed within an existing mobile care program (like CareMother), which includes home-based and centre-based antenatal tests via a portable kit and mobile application, used by a health worker for real-time transfer of results to doctors. Alternatively, Fetosense can be deployed as a solution within existing maternal care programs in Bangladesh. 

Fetosense will be comprehensive enough to enable frontline workers to take informed decisions and refer mothers for timely intervention, to eventually improve neonatal outcomes. Fetosense is unaffected by maternal obesity and movements. It is based on Ultrasound doppler, where studies have shown that changes in variability indices like accelerations, decelerations and beat-to-beat variability significantly correlate to fetal distress (Jezewski et al, 2006).

At the Bottom of Pyramid, Fetosense will be adopted through its easy-to-use configuration and interface, portability and linkage with a smartphone app, by doctors at primary health centres, midwives and 'close-to-client' community health workers. Fetosense will report transparent real-time data for effective clinical decisions – for e.g. abnormal beat-to-beat variability could be linked to fetal hypoxia and its prediction will be flagged on the mobile application. This prediction will be based on the analysis of a large sample of ‘training data’ that we are currently performing. Access to thousands of mothers under our CareMother program will allow such an analysis. Fetosense offers the aforementioned strengths at 33% lesser cost than cardiotocography machines.

Fetosense is an AI-based, non-invasive, wireless and portable, smartphone integrated and affordable fetal heart rate (FHR) and distress monitor. We have developed an auto-interpretation system of FHR records (using Dawes-Redman criteria) and real-time remote monitoring to empower doctors/nurses/health workers to measure fetal cardiac function and its consequences on the fetus. Thus, health staff can take informed decisions and refer mothers in time for intervention, to improve neonatal outcomes. Fetosense is unaffected by maternal obesity and movements, and overcomes the limitations of cardiotocography (CTG) devices. A health worker can place the belt on the abdomen and start the device without repositioning. Fetosense is based on the principle of Ultrasound doppler, which significantly correlates to fetal distress using variability indices like accelerations, decelerations and beat-to-beat variability (Jezewski et al, 2006). 

At the grassroot level, Fetosense will be adopted because:

1. It enables AI-based first-level immediate interpretations and reduces the dependence on a gynecologist for a preliminary diagnosis; this 'triages' a large number of patients
2. It enables real-time remote montoring to receive the gynecologist's advice during needy hours
3. It has smartphone-based portability and targets ‘door-step’ community health workers, nurses and midwives, close to the mothers
4. Physicians can leverage the per-test based revenue sharing model that reduces the test cost by half for the patient (compared to gynaecologist-operated CTG). Moreover, under the Government model, this cost can be subsidized or made free for the beneficiaries.

Therefore, Fetosense is an example of a renewed application of an existing technology, and a change in the process and business model to increase the solution's effectiveness for pregnant women.

Fetosene was validated against the standard of care Non-Stress Test (NST) device i.e. cardiotocography or Toco, in October 2017, at a tertiary hospital in Mumbai, India, in a sample of 50 pregnancies, under the supervision of a world-renowned obstetrician and gynaecologist with over 20 years of experience in NST interpretation. The sample diversity was maintained as per statistical requirements by considering sample distribution over a gestational age range (24-40 weeks), maternal weight classes (48-102 kg), maternal age (21-40 years) and meeting of high-risk criteria (10% of mothers). For each mother, fetal heart rate (FHR) was recorded with Toco and Fetosense, one after another, for a duration of 10-20 minutes each, plotted over a 1 cm/minute scale. Results verified by the doctor were as follows: a) Accuracy: Average baseline FHR was non-significantly different between Fetosense and Toco (p=0.0697); Fetosense reported a sensitivity of 94% in the assessment of FHR; b) Clinician ease: Excellent clinical interpretability in 40% of cases compared to Toco; c) Logistics: Fetosense has end-to-end wireless configuration, it is easy to carry and set-up, and printing graphs through the wireless A-4 printer is convenient (compared to the more expensive thermal paper), it has robust device quality and easy configuration; d) Maintenance: Fetosense can work for ~6 hours when fully charged to perform ~40 NST tests in a single charging cycle.

Moreover, Fetosense will benefit out of the prior work under our CareMother program since 2015 viz. a scale-up environment, and program learnings through experiences of health workers and stakeholders.

1. Through Fetosense, we are currently serving 400 pregant women since the launch of the solution in June 2019

2. We will be serving 3000 patients in year-1 through a commercial pilot (deployment of 30 devices)

3. We aim to serve 3 million pregnancies in 5 years based on our prior research of market size and our go-to-market strategy (deployment of 10,000 devices)

Within the next year, we aim to undertake a hospital- or clinics-based validation study in urban and rural settings in Bangladesh to replicate the results obtained in our prior validation in India. In parallel, we aim to complete the regulatory processes for enabling expansion of fetal monitoring services in Bangladesh.

Over a five-year period, we have the following goals:

1. Undertake a two-year multi-centric (rural and urban) commercial pilot covering 14,000 pregnancies by deployment of 110 Fetosense devices, with its revenue-sharing business model, and leverage the strength of doctors, nurses and community health workers (CHWs) with remote monitoring by specialists. It should be noted that CHWs in Bangladesh have been increasing in size since the nineties (current: 14 per 10,000 people, Save the Children data 2011), and are cost-effective (Ahmed SM et al, 2011).

2. Collaborate with the 1646 members of the Obstetrical and Gynaecological Society of Bangladesh (OGSB) to create knowledge partnerships with opinion leaders and establish a national-level Fetosense Mentorship Program of doctors, nurses, midwives and CHWs. This program will facilitate the training/mentoring of frontline health workers by the specialists, and form effective referral linkages and remote monitoring systems via the expanded usage of Fetosense.

3. Leverage the results of the aforesaid pilot and collaborations to expand Fetosense's AI-based services to the majority of frontline health workers by year-5, and forge partnerships with regional and/or central Government agencies to achieve the same.

4. Develop partnerships with companies having expertise in supply chain and distribution for mitigating scale-up barriers

Challenges for the next year:

1. Regulatory Challenges: Medical devices can face regulatory challenges especially in cross-country settings and for performing validation studies. Necessary approvals need to be in place.

Challenges for the next five years:

1. Market Challenges: Supply chain and distribution challenges will be pronounced in case of medical devices; brand building is a lengthy process and especially challenging in the medical field. Establishing distribution and support system with the per-test based revenue-sharing model will be important and herein, aligning the incentives of each party and maintaining quality of service for brand building will be crucial.

2. Operational Challenges: Queries by health staff of varying skill-sets using Fetosense on-the-ground will be frequent and would need to be addressed promptly, even though the solution is easy to use and has been customized for their usage. Fetosense users will also need to factor device calibration-related issues. Linguistic barriers will need to be resolved.

3. Financial Challenges: Sufficient data would need to be obtained on procurement and development costs and recurring material costs.

1. Regulatory Challenges:

We have adhered to physical and metrological standardizations for Fetosense in India, by comparing its output with a previously approved similar device. We will obtain additional third-party validations from authorized agencies at the field-level. So far, we have obtained CE certification and we follow HIPPA, FHIR, ISO and NHM's health data privacy, storage and inter-operability standards. We will take similar efforts to meet the regulatory needs in Bangladesh.

2. Market Challenges:

We have gained learnings from our partnership with the Federation of Obstetrical and Gynaecological Societies of India (FOGSI) and received their support for advocacy and exhibition exposure at conferences. We will follow a similar approach with the federation in Bangladesh. To mitigate supply chain and distribution challenges, we will partner with medical device and pharma companies like GE (Genworks), Philips, J&J, Zydus etc. We will promote doctor-targeted social media broadcasting for brand building.

3. Operational Challenges:

With a multi-lingual (Bengali, English) user manual, effective training, dedicated central team's monitoring and a helpline number, the health staff will be able to resolve any contingencies. For challenges arising out of new test configuration, the software will be remotely upgraded on the phone. In terms of calibration, Fetosense's self-calibration will help, as well as our continual effort to improve its accuracy relative to past performance (via back-end analytics).

4. Financial Challenges:

There is no direct anticipated financial risk due to a defined budget and our current funding sources. Procurement and development costs and recurring material costs have been tested during prior validations.

There are an estimated 5 million pregnancies annually in Bangladesh. Given the need to perform fetal monitoring in both normal (at least twice during gestation) and high-risk pregnancies (at least four times during gestation), an estimated 17 million tests can be made available using Fetosense every year. Thus, the total addressable market size in 5 years, assuming $2 per test (under revenue-sharing model) is $170 million. We will provisionally aim for achieving 2% of this market size, which is $3 million, 1.5 million tests and at least 400,000 pregnancies. 

20 full-time staff

4 part-time staff

Under the CareMother program, we have partnered with 15+ implementing organizations (Government agencies, non-profits and private hospitals) in ten states of India.

We have three researchers on-board who are developing AI-based capabilities in Fetosense as well as a dedicated hardware developer and biomedical research engineer.

The project lead, Dr. Ameya Bondre is a Master of Science in Public Health graduate from Johns Hopkins University, Baltimore, with prior medical training from KEM Hospital, Mumbai. He is a Global Health Field Placement fellow at Hopkins with more than 8 years of work experience with trials, cross-sectional studies and qualitative research in diverse settings in India and Africa, including in antenatal and postnatal health. He has clinical experience in urban and rural facilities. He is specialized in quantitative (Stata and SPSS) and qualitative data analysis (Nvivo) and training and capacity-building of field workers. With strong writing and communication skills, he has several peer-reviewed publications. He is presently heading the clinical research and development division of CareNX Innovations. 

The team is strong and inter-disciplinary with expertise in software development, AI (three researchers), business development, embedded hardware development and operations. CareNX Innovations has mentors in Stanford University, MIT-D Lab and IIT-Bombay in addition to mentoring from experienced gynaecologists based in Mumbai, India and abroad. 

On the whole, our diverse skill-profile, current partnerships and long-standing experience in delivering a mobile maternal care program (CareMother) in ten states of India, uniquely position us to deliver Fetosense to the large majority of mothers in Bangladesh.

CareNX Innovations, the parent company for Fetosense, is associated with IIT-Bombay as its incubating organzation. The Society for Innovation & Entrepreneurship (SINE) has been supporting the work of CareNX Innovations since 2015. Our key external channel for technical support is Dr. P.G. Natarajan (with 20+ years of his clinical experience), at Nanavati Hospital, where we had conducted our first hospital pilot. We have also collaborated with the Federation of Obstetrical and Gynaecological Societies of India (FOGSI), who is our knowledge partner for advocacy, device promotion and exhibition exposure at conferences.

Importantly, Fetosense has received funding and mentoring from CISCO Global Problem Solver Challenge, Department of Science and Technology - Government of India, and Google Launchpad Accelerator.

Business models for Fetosense:

1. B2B is a revenue-sharing model between the company and the doctor with monthly invoicing on the basis of a per-test cost of ~$4 ($2 to be paid to the company by the doctor). We will commercially pilot the device over two years and deploy 110 devices in a phased manner, to eventually serve 14,000 pregnancies. We will achieve the return-of-investment (RoI) at ten months after deployment. We worked out the RoI based on capital and operational expenditure for one device for one year and the number of tests to be done over a year through a device. When the test is physician-based, the cost of a gynaecologist's consultation gets deducted from the total cost. The estimated price (based on our work in India) in the absence of a revenue-sharing AI-based model is $8 (borne by the patient). AI allows the physician or health worker to make an auto-interpreted first-line diagnosis that he/she can communicate to the patient for an additional, but affordable cost. This cost of $4 will be further subsidized for patients under Government programs (e.g. with health workers/midwives).

2. CareNX Innovations can establish annual contracts with local municipal governments or public hospitals for revenue sharing model (B2B) with minimum 500 tests and 10% subscription fees, to cover 90% of the capital investment on the device within the first month of deployment. Subsequent operational cost will get covered given the high volume of tests at these facilities.

There is a clear need for a device like Fetosense for health staff in low income settings. Utility of AI-based first-level interpretations and revenue sharing commercial model brings down the total test cost. With this scalable model, Fetosense will sustain in the following ways:

1. Return-of-investment (RoI) and recurring revenue (50%): RoI of revenue sharing based model with initial 110 devices (two-year pilot, 14000 pregnancies) is estimated to be 10 months for CareNX Innovations. Revenue after this period will generate profits to take care of on-going operations. These profits will be used to deploy more Fetosense devices in the same market.

2. Annual revenue contracts (20%): CareNX will establish annual contracts with local municipal governments or public hospitals for revenue sharing model with minimum 500 tests and 10% subscription fees, to cover 90% of the capital investment on the device (made by the company) within the first month from deployment. Subsequent operational cost will get covered by the high volume of tests at these facilities.

3. Debts/Grants (30%): CareNX received funds earlier from Grand Challenges Canada and CISCO Global Problem Solver, and it continues to apply for grants to generate the capital needed for commercial scale-up of the revenue sharing model through channel partners and distributors. 

We are applying to the Tiger IT challenge for the following key reasons and motivations:

1. There is sufficient similarity between the local maternal and child health context in Bangladesh to that in India where we have ground-level experience of providing services, and that makes Bangladesh a strategic region to further our work in fetal monitoring and protecting newborn lives.

2. Cross-country learnings between our work in India and Bangladesh will be of immense value to achieve the goals of national-level programs in maternal and child health.

3. Fetosense has been designed for settings similar to those in India and Bangladesh, for instance, the nature of its design, ease of usage, and compatibility with the skill-set of a frontline worker, the scarcity of gynaecologists in both countries and the necessity of remote-monitoring, and the need for AI-based early diagnosis and management to reduce fetal and neonatal complications and address the 'resistant' causes of neonatal mortality, which have roots in antenatal care.

4. Tiger IT's funding and other strategic and operational support from a network of seasoned professionals and entrepreneurs in Bangladesh, aligns with our company's investment goals and financial growth plans.

We would like to partner with the following organizations based in Bangladesh:

1. The Obstetrical and Gynaecological Society of Bangladesh

For reaching out to specialists and facilitating mentorship of physicians, nurses, midwives and health workers; also for advocacy of the solution

2. Key non-profit organizations to develop local partnerships with communities and health systems to enable effective expansion of services. An example is Bangladesh Rural Advancement Committee (BRAC) 

3. Major private hospitals in urban Bangladesh

4. Government health agencies for greater expansion of fetal monitoring services and their integration within regional and national health programs