HealthPulse AI Leverages MediaPipe to Improve Well being Fairness


A visitor submit by Rouella Mendonca, AI Product Lead and Matt Brown, Machine Studying Engineer at Audere

Please observe that the knowledge, makes use of, and purposes expressed within the under submit are solely these of our visitor authors from Audere.

About HealthPulse AI and its software in the true world

Preventable and treatable ailments like HIV, COVID-19, and malaria infect ~12 million per yr globally with a disproportionate variety of instances impacting already underserved and under-resourced communities1. Communicable and non-communicable ailments are impeding human improvement by their detrimental impression on training, revenue, life expectancy, and different well being indicators2. Lack of entry to well timed, correct, and reasonably priced diagnostics and care is a key contributor to excessive mortality charges.

As a consequence of their low value and relative ease of use, ~1 billion speedy diagnostic checks (RDTs) are used globally per yr and rising. Nevertheless, there are challenges with RDT use.

  • The place RDT information is reported, outcomes are laborious to belief on account of inflated case counts, lack of reported anticipated seasonal fluctuations, and non-adherence to therapy regimens.
  • They’re utilized in decentralized care settings by these with restricted or no coaching, growing the danger of misadministration and misinterpretation of check outcomes.

HealthPulse AI, developed by a digital well being non-profit Audere, leverages MediaPipe to handle these points by offering digital constructing blocks to extend belief on the earth’s most generally used RDTs.

HealthPulse AI is a set of constructing blocks that may flip any digital resolution right into a Speedy Diagnostic Take a look at (RDT) reader. These constructing blocks clear up outstanding international well being issues by bettering speedy diagnostic check accuracy, decreasing misadministration of checks, and increasing the supply of testing for situations together with malaria, COVID, and HIV in decentralized care settings. With only a low-end smartphone, HealthPulse AI improves the accuracy of speedy diagnostic check outcomes whereas robotically digitizing information for surveillance, program reporting, and check validation. It gives AI facilitated digital seize and end result interpretation; high quality, accessible digital use directions for supplier and self-tests; and requirements based mostly real-time reporting of check outcomes.

These capabilities can be found to native implementers, international NGOs, governments, and personal sector pharmacies by way of an internet service to be used with chatbots, apps or server implementations; a cellular SDK for offline use in any cellular software; or straight by native Android and iOS apps.

It permits progressive use instances reminiscent of quality-assured digital care fashions which permits stigma-free, handy HIV dwelling testing with linkage to training, prevention, and therapy choices.

HealthPulse AI Use Instances

HealthPulse AI can considerably democratize entry to well timed, high quality care within the personal sector (e.g. pharmacies), within the public sector (e.g. clinics), in group applications (e.g. group well being employees), and self-testing use instances. Utilizing solely an RDT picture captured on a low-end smartphone, HealthPulse AI can energy digital care fashions by offering invaluable determination assist and high quality management to clinicians, particularly in instances the place traces could also be faint and laborious to detect with the human eye. Within the personal sector, it may possibly automate and scale incentive applications so auditors solely have to assessment automated alerts based mostly on check anomalies; procedures which presently require human critiques of every incoming picture and transaction. In group care applications, HealthPulse AI can be utilized as a coaching instrument for well being employees studying find out how to accurately administer and interpret checks. Within the public sector, it may possibly strengthen surveillance methods with real-time illness monitoring and verification of outcomes throughout all channels the place care is delivered – enabling sooner response and pandemic preparedness3.

HealthPulse AI algorithms

HealthPulse AI gives a library of AI algorithms for the highest RDTs for malaria, HIV, and COVID. Every algorithm is a set of Pc Imaginative and prescient (CV) fashions which can be educated utilizing machine studying (ML) algorithms. From a picture of an RDT, our algorithms can:

  • Flag picture high quality points widespread on low-end telephones (blurriness, over/underexposure)
  • Detect the RDT kind
  • Interpret the check end result

Picture High quality Assurance

When capturing a picture of an RDT, it is very important make sure that the picture captured is human and AI interpretable to energy the use instances described above. Picture high quality points are widespread, significantly when photographs are captured with low-end telephones in settings that will have poor lighting or just captured by customers with shaky fingers. As such, HealthPulse AI gives picture high quality assurance (IQA) to determine adversarial picture situations. IQA returns issues detected and can be utilized to request customers to retake the picture in actual time. With out IQA, shoppers must retest on account of uninterpretable photographs and expired RDT learn home windows in telehealth use instances, for instance. With just-in-time high quality concern flagging, extra value and therapy delays could be prevented. Examples of some adversarial photographs that IQA would flag are proven in Determine 1 under.

Images of malaria, HIV and COVID tests that are dark, blurry, too bright, and too small.Determine 1: Photos of malaria, HIV and COVID checks which can be darkish, blurry, too vibrant, and too small.


With simply a picture captured on a 5MP digicam from low-end smartphones generally utilized in Africa, SE Asia, and Latin America the place a disproportionate illness burden exists, HealthPulse AI can determine a selected check (model, illness), particular person check traces, and supply an interpretation of the check. Our present library of AI algorithms helps lots of the mostly used RDTs for malaria, HIV, and COVID-19 which can be W.H.O. pre-qualified. Our AI is situation agnostic and could be simply prolonged to assist any RDT for a spread of communicable and non-communicable ailments (Diabetes, Influenza, Tuberculosis, Being pregnant, STIs and extra).

HealthPulse AI is ready to detect the kind of RDT within the picture (for supported RDTs that the mannequin was educated for), detect the presence of traces, and return a classification for the actual check (e.g. constructive, detrimental, invalid, uninterpretable). See Determine 2.

Figure 2: Interpretation of a supported lateral flow rapid test.Determine 2: Interpretation of a supported lateral circulate speedy check.

How and why we use MediaPipe

Deploying HealthPulse AI in decentralized care settings with unstable infrastructure comes with various challenges. The primary problem is an absence of dependable web connectivity, typically requiring our CV and ML algorithms to run domestically. Secondly, telephones accessible in these settings are sometimes very previous, missing the most recent {hardware} (< 1 GB of ram and comparable CPU specs), and on totally different platforms and variations ( iOS, Android, Huawei; very previous variations – presumably now not receiving OS updates) cellular platforms. This necessitates having a platform agnostic, extremely environment friendly inference engine. MediaPipe’s out-of-the-box multi-platform assist for image-focused machine studying processes makes it environment friendly to fulfill these wants.

As a non-profit working in cost-recovery mode, it was essential that options:

  • have broad attain globally,
  • are low-lift to keep up, and
  • meet the wants of our goal inhabitants for offline, low useful resource, performant use.

Without having to jot down numerous glue code, HealthPulse AI can assist Android, iOS, and cloud gadgets utilizing the identical library constructed on MediaPipe.

Our pipeline

MediaPipe’s graph definitions enable us to construct and iterate our inference pipeline on the fly. After a person submits an image, the pipeline determines the RDT kind, and makes an attempt to categorise the check end result by passing the detected result-window crop of the RDT picture to our classifier.

For good human and AI interpretability, it is very important have good high quality photographs. Nevertheless, enter photographs to the pipeline have a excessive degree of variability we now have little to no management over. Variability elements embrace (however are usually not restricted to) various picture high quality on account of a spread of smartphone digicam options/megapixels/bodily defects, decentralized testing settings which embrace differing and non-ideal lighting situations, random orientations of the RDT cassettes, blurry and unfocused photographs, partial RDT photographs, and lots of different adversarial situations that add challenges for the AI. As such, an essential a part of our resolution is picture high quality assurance. Every picture passes by various calculators geared in direction of highlighting high quality issues that will forestall the detector or classifier from doing its job precisely. The pipeline elevates these issues to the host software, so an end-user could be requested in real-time to retake a photograph when vital. Since RDT outcomes have a restricted validity time (e.g. a time window specified by the RDT producer for a way lengthy after processing a end result could be precisely learn), IQA is important to make sure well timed care and save prices. A excessive degree flowchart of the pipeline is proven under in Determine 3.

Figure 3: HealthPulse AI pipelineDetermine 3: HealthPulse AI pipeline


HealthPulse AI is designed to enhance the standard and richness of testing applications and information in underserved communities which can be disproportionately impacted by preventable communicable and non-communicable ailments.

In direction of this mission, MediaPipe performs a essential function by offering a platform that permits Audere to shortly iterate and assist new speedy diagnostic checks. That is crucial as new speedy checks come to market repeatedly, and check availability for group and residential use can change regularly. Moreover, the pliability permits for decrease overhead in sustaining the pipeline, which is essential for cost-effective operations. This, in flip, reduces the price of use for governments and organizations globally that present companies to individuals who want them most.

HealthPulse AI choices enable organizations and governments to profit from new improvements within the diagnostics house with minimal overhead. That is a vital part of the first well being journey – to make sure that populations in under-resourced communities have entry to well timed, cost-effective, and efficacious care.

About Audere

Audere is a world digital well being nonprofit creating AI based mostly options to handle essential issues in well being supply by offering progressive, scalable, interconnected instruments to advance well being fairness in underserved communities worldwide. We function on the distinctive intersection of world well being and excessive tech, creating superior, accessible software program that revolutionizes the detection, prevention, and therapy of ailments — reminiscent of malaria, COVID-19, and HIV. Our numerous staff of passionate, progressive minds combines human-centered design, smartphone expertise, synthetic intelligence (AI), open requirements, and the perfect of cloud-based companies to empower innovators globally to ship healthcare in new methods in low-and-middle revenue settings. Audere operates primarily in Africa with initiatives in Nigeria, Kenya, Côte d’Ivoire, Benin, Uganda, Zambia, South Africa, and Ethiopia.

1 WHO malaria reality sheets

Supply hyperlink

You might also like