AIMS - Artificial Intelligence for Multiple Sclerosis

AIMS is a mobile health platform directed towards reducing limitations in diagnosing and monitoring MS worldwide by utilizing Bayesian networks to more accurately correlate symptom patterns with neurological disorders. If implemented, AIMS could serve as a simple aid for MS detection and surveillance accessible to anyone with a smart device.

There is currently a significant lack of accurate diagnostic tools for MS due to gaps in knowledge regarding its etiology. In 2015, the MS Society found that 4 out of every 5 MS patients in the UK are misdiagnosed at least once, often waiting years between the onset of their symptoms and their diagnosis. Even worse, symptoms used to detect MS are found in other similar diseases such as myasthenia gravis and sarcoidosis, and change drastically in a patient over time. Accordingly, symptom patterns must be established to properly diagnose and monitor the progression of MS for optimized treatment.  

While symptoms are one of the primary aspects of diagnosing MS, they are highly variable from patient to patient and can appear sporadically for years before visible disability sets in. AIMS would solve the mounting problem of misdiagnosis by establishing patterns of these symptoms in the beginning stages of MS through using a Bayesian network to mine data of consenting users. The network would aid neurologists by generating maps of patient data over time that could scale the probability of disease presence and alert users of potential conditions they might have with warning signals sent to their mobile devices. 

The intended result of AIMS is to lower the amount of misdiagnosed cases of MS annually by serving as a personalized medical resource available anywhere. More specifically, this app would benefit physician researchers at clinics with large volumes of patients and users who come from both developed and developing nations. Because the platform would be implemented through an app, it would be accessible to a demographically diverse audience comprised of users from many countries and socioeconomic backgrounds. As a result, the impact of AIMS would include ameliorating diagnostic standards for MS and decreasing health disparities for underserved populations. 

Ten hospitals and universities will be selected to participate in beta testing to refine app performance and build a user base - Build partnerships with neurology clinics to grow app audience

Scalable pattern mining will be employed to distinguish the presence of MS from other diseases by analyzing symptoms - Discover symptom patterns in MS patients using a Bayesian network

Establish a robust social media presence to reach people from disadvantaged backgrounds - Decrease health disparities by providing a medical platform for anyone with a smart device

Machine learning tools such as artificial neural networks have long been used in the field of medical diagnostics to detect disease pathology with astounding accuracy. Direct access to such innovations, however, still remains a noticeably latent option for the general public. By combining probability prediction networks with data mining, AIMS will serve as a unique healthcare resource that will enable the general public make use of such technology by downloading an app with a user-friendly interface. While symptoms are typically used to gain often falsely conclusive diagnoses of a diseases, AIMS will change this by instead scaling disease probability. 

The purpose of AIMS is to facilitate personalized neurological healthcare for people regardless of their geographic location. Historically, systems that have disseminated access to medical resources have marginalized the most vulnerable populations on the basis of economic constraints. By creating an app that can be downloaded by anyone with a smart device, the technology in AIMS will fundamentally burgeon public access to neurological information in a user-friendly way, as the majority of the world's population use some form of mobile technology. In doing this, AIMS would enable underserved populations to obtain health information they would otherwise not have. 

To reach a broad audience in a practical fashion, AIMS will be dispersed to patients via a mobile app available on the App Store and Google Play. In 2013, the UN calculated that 6 out of 7 people use mobile devices globally, making apps the ideal mode for this technology. Underserved patient populations will be able to access the app as it will be priced at $1.99, with no in-app purchases required to take advantage of its full features. Hospitals would be able to setup doctor-patient communication networks in the app for a monthly fee of $4.99. 

Operational funds for AIMS will be obtained in three different forms:

1. Equity crowdfunding (Research stage)

Initially, online crowdsourcing platforms will be used to secure a sum of $5000 to formally setup AIMS. This money could help with the legal fees associated with founding a startup and would pay for an LLC license. 

2. Research grants (Research and Pilot stages)

After incorporation as an LLC, AIMS will acquire funds for conducting a pilot study at a neurology clinic by receiving grants that would be used to compensate participants for their time. The required sum of the grant would need to be between $1000 - $5000. 

3. Seed capital (Growth and Scale stages)

Equity stakes in the company (<25%) will be exchanged for money from investors that will cover operational costs and help boost the online presence of AIMS by paying for ad space and a website designer.

Financial constraints have been the biggest hindrance to the development of this solution. Although the software components of AIMS require little to no outside equity, it is much more difficult to test out the resulting networks on actual MS patients without having any funds secured to conduct preliminary studies. Thus, the absence of venture capital is the primary issue holding back the progress of this enterprise. 

AIMS has the potential to refine both the understanding and diagnostic standards of MS in its early stages through harnessing the largely unrealized potential of combining machine learning with mobile technology. Without the proper connections and support, however, it is difficult to develop this solution. Through becoming a Solver, I hope to change this by building the partnerships necessary to implement this vision on a global scale by showcasing my vision to MIT and networking with fellow Solvers. Personalizing neurological care is a daunting task, but a challenge that I believe I can take on with the Solve community.

AIMS currently has no financial partnerships, but does receive other forms of support from a research lab at a major university and a national non-profit organization. 

Healthtap, Moving Analytics, Vital Connect