Development of web applications for self-diagnosis

Development of web applications for self-diagnosis

February 9, 2016

You only need to type ‘online diagnosis’ into any search engine to come up against a long list of web applications which will supposedly help you to identify an illness. Nowhere, however, will you find enthusiastic reviews of a particular website that has actually given an accurate diagnosis. Why so? Well, let's take a look at the main difficulties of attempting to create an online diagnostics tool.

The question of motivation

By analysing the diagnostics tools that we can find online, we might come to the conclusion that there aren't many developers actually devoted to solving real issues for their users. The purpose of most of these applications is to drive traffic and, ultimately, to monetise the project. There are also aspects developed as PR resources to improve a clinic's image.

The usefulness of such medical web applications is judged first and foremost by the question of whether the program can answer the user's questions: What is the problem? What kind of doctor should the patient see? What tests should be carried out? How dangerous is the issue at hand? Let's suppose that on the basis of the location of a headache, the online diagnostics tool has come to the conclusion that there is a 25% chance of hypertension and a 10% chance of a migraine without aura. Would this make you feel better? These are the results taken from a real algorithm of one of the best diagnostic tools out there.

The main issue with online diagnostics tools is that the developers very rarely ask themselves which problems their project is supposed to solve. They are not interested in this, believing that there are enough hypochondriacs in the world to bring traffic to the site. It doesn't matter that upon clicking the link, they find nothing more than a glorified toy. After all, only a qualified doctor can provide an adequate diagnosis and administer proper treatment, so why bother?

At the same time, to create something useful, one has to deal with the following questions:

  • How can we help patients before they go to the doctor?
  • How can we help when a doctor is unavailable for a consultation?
  • What reasons might prevent access to medical assistance?

This information should serve as the starting point, before one starts to design the project.

Stupidity is worse than treason

It is understandable why a programmer should endeavour to simplify an application; the less the user has to click the mouse with furrowed brows, trying to identify the characteristics of a skin rash, the better. This has led to a problem common to most applications, whereby the algorithm attempts a diagnosis on the basis of just one or two symptoms. A doctor, on the contrary, looks at a combination of symptoms when making a diagnosis, not to mention that the symptoms themselves have certain nuances to be kept in mind. There isn't just pain or the presence of a rash. Discharge has a source, colour, consistency and smell, all of which needs to be taken into account.

It is impossible to create a truly effective software which turns an electronic medical reference book into an interactive experience. There is no way to design an algorithm for diagnostics which works independently of a doctor (a person who actually knows what they are doing). Simplicity, therefore, has to be sacrificed, and we can provide it only on the level of interface.

When can the application help?

Let's return to the question of usefulness. It stands to reason that a doctor is indispensable in the diagnosis and the prescription processes. However, the patient doesn't only face the question of how to treat the symptoms. Before meeting a doctor, the patient should try to eliminate the main symptoms. It wouldn't do any harm to find some over-the-counter medicine to alleviate the condition. For instance, if a patient has aching joints, a prompt to use Ibuprofen instead of Analgin would be appropriate.

In such situations, first aid instructions would be of great use, especially when the problem is not a particularly common one. You will rarely find information on administering emergency care to someone having a heart attack or stroke on TV or in medical brochures. Such applications might also help when a person is unable to decide how to act. Should they call an ambulance or just rest in bed to convalesce? What kind of doctor is required? Sometimes, one might simply need seemingly obvious information, such as an ambulance's phone number, which will vary depending on the region and network provider, and can often be hard to find.

A web application providing the best possible answers to patient queries would be the most popular. As is clear, from all that has been outlined, this will need serious investment.

Why undertake it?

Primary web diagnostics will be helpful to clinics which attract patients online. If a patient gets answers to some questions before seeing a specialist, they will tend to be more receptive to a doctor who is able to clarify the situation. However, a private medical company, as a rule, is not ready to spend the money needed to create a truly useful web application with a multistage diagnostic algorithm and professional recommendations.

At the same time, a quality web application for self-diagnosis would see great demand from ordinary users, since opting for a healthy way of life and control over one's body is becoming a huge trend. Whoever is the first to enter the market as an online diagnostics professional, capable of providing real help to people, has an opportunity not only to capture the market, but also to receive state support, since the problem of interactive diagnostics in public healthcare also needs to be developed.