This could be an good idea for people who want to start a business which serves a noble cause too, specifically in the domain of Healthcare. So the basic idea is to provide an asynchronus clinical diagnostic decision support system (CDSS) to healthcare professional on a hand-held device like (tablet/pda/notbook) or a desktop . The way CDSS works is, we have a very big number of cases and expert opinions combined to form rules for diagnosing what disease the next patient (case) has an the basis of symptoms seen. Most systems use bayesian algorithms to formulate these rules. When a new case comes, on the basis of information on symptoms observed and other relevant personal entered, system suggests what diseases could possible be there with probabilities for each of these and explanations for why that disease could be present. It would also suggest what tests should be done to make a diagnosis and in what sequence. Finally the healthcare professional (most often a doctor) makes a decision or diagnosis and treats the patient for that disease. Ideally each new case with final test results and the diagnosis decision should be updated in the system so that rules also get updated for treating the next set of patients better.
Additional support for treatment guidance and sending reminders for taking the medication or check-ups etc can also be provided. Use of Information and Communication Technologies (ICT) in healthcare is increasing but has not been fast enough to catch up with the pace at which technologies are developing. Technically, building such a system is really not a big deal but there is a lack of effective delivery mechanisms which gets adopted well into the healthcare community which apparently is normally not well-versed in use of ICTs. Right now such systems come along with hospital information systems and are thought of as something big, complicated and expensive. What I think is, we can just use Google Gears (http://gears.google.com ) as a platform application to build up our own asynchronus application which can run on any client and updates the rules by periodically sending the information/data about the cases treated in the last unconnected period and also downloading the information about updates in the diagnosis rules. This way we can provide a good product at affordable prices to the users. Google provides APIs for using its applications to anyone, and that is a great resource which many entrepreneurs may want to exploit for delivering their services.
The amount of information being generated in healthcare is so huge that it is almost impossible for professionals to keep up with all the latest information in the field. Such systems are an answer to it, as such this is something people have known since long but now is the best timing for launching such systems as healthcare professional are increasingly adopting ICT based systems into their practice. Points mentioned so far hold true for the urban and developed areas around the world and for the underdeveloped areas too. But for underdeveloped and remote (rural) areas where doctors are rarely available this application is going to be far more beneficial, if we can provide healtcare advice through telemedicine using such decision support systems. It is an emerging area and a huge market around the globe. Small entrepreneurs can also start such a system, and if such small entrepreneurs can become the part of a bigger network (something like a cooperative) for sharing the data collected, then it would definitely be called as a great innovation. I believe one group trying to something very similar thing is www.d-tree.org and I am certain that there would be many more small businesses or non-profits working on this idea.
The first part of such a business would be to collect the information or cases and develop a system which lies on the server. This is possible with some investment of time and money and many firms can do it so to being with an alliance can also be formed. There are numerous initiatives at different universities which have already have a lot of data with them, and I believe one can either partner with them or just buy the data from them. The second and most crucial part would be to efficiently provide such a decision support to the practitioners or telemedicine providers in user friendly way. Direct diagnostic advice to consumers is out of question right now because of the legal issues involved but as some legal barriers fall in future due to increased trust in technology, major players would be able to take advantage of those new laws too. Existing APIs of Google Gears can provide the web in a condensed form asynchronously to client devices, and hence have most of the features which would be required to implement a health decision support in the form proposed above.
Another orthogonal but important aspect here is that governments and insurance providers have a huge interest in reducing healthcare costs and increase efficiency of the healthcare in general. With support of such a system a lot of spurious diagnosis testing could be avoided, because that happens due to either some personal interests of doctors, hospitals and testing labs, or because of the over cautious approach towards diagnosis adopted by the doctors and healthcare providers. These spurious tests account for about a third of total lab tests done specially in the developed economies. So we have stakeholders in the market who have a strong interest in promoting such products and may be making these mandatory in future. New breed of healthcare professionals are receptive to the idea of using ICTs for providing better healthcare and other complementary products, systems or ideas like electronic health records for individuals are also getting acceptance into the market. On the whole use of ICTs in healthcare is gaining momentum and this kind of product/service to deliver diagnosis support would be one of the pillars to provide healthcare in a networked manner.
I believe, it is a good idea for anyone who wants to make money and also do something meaningful. Also, everyone is aware of the humongous size of healthcare market. If anyone would like to discuss more details on this then I would be more than glad to so :-)