Eight years ago, my parents were affected by glaucoma. My mother was prescribed topical drops but they stung her eyes, so she stopped taking them, and even omitted to tell the physician about her situation. As a result she lost her eyesight in one eye. Around the same time, Enrique Malaret, our COO and co-founder of Injectsense, had also come up against the disease. We decided we would do what we could to help. The challenge spoke to us as individuals, as much as engineers. How do we determine whether a treatment actually works? Our solution was a continuous monitoring device, taking the form of an injectable ultra-miniature sensor coupled with a secure digital health platform. We spent the next three years fleshing out the concept and securing funding – once we had enough, we began to scale.
As you can imagine, creating an organ-to-cloud system is not easy. It takes time to develop a premium product, especially one that draws from several domains – such as medicine, semiconductors, fluid interactions, and electronics. Putting together a team that could connect these disciplines without leaving gaps was critical. We made sure to pick people with overlapping skills, and ensured we worked as one collective entity. The team now stands at 14-plus employees, alongside a pool of contractors.
We choose to keep our team small because it keeps us agile. It also makes it easier for everybody to know their roles, which is important, as two-thirds of the start-ups fail because of internal conflict. We made a policy of looking for people who are hands-on, team players, with proven multidisciplinary expertise, and who are motivated well beyond simple monetary payback. We have stuck to it. Together, we have built a team that has developed a solution to be proud of.
So how does it work? First, the implant is delivered in-office via an injection, leaving a self-anchoring sensor in the vitreous base. The implant begins to measure IOP at a series of times predetermined by the physician. Once a week, the patient is reminded to upload the data by putting on a pair of glasses – this notification will come from an app on the patient’s phone. Uploading takes less than a second. At this point, the data is sent to the cloud, where it is ready to be mined by our software. We start by removing data that has no purpose (such as IOP increases caused by blinking) and instead, look for information that is actionable, which is to say, pressure readings that exceed a certain value over a given period of time.
The idea is to build a histogram representative of the pressure exposure. The physician can then access the data and see which patients, if any, have exceeded their set target. We also provide physicians with the option to decide how often patients are sampled – for example, once every ten minutes or once an hour. The next time the patient uploads data via the glasses to the cloud, the new sequence will be automatically downloaded.
The whole system requires very little effort from the patient. Minimizing human error was one of our core aims, as patients are often the weakest link in the treatment process. Every physician can provide an example of a case where patient non-compliance has resulted in less than ideal outcomes. This issue is particularly common in glaucoma, where patients do not necessarily see an immediate benefit in taking their medication but may experience side effects. The patients also may not see the urgency, as they don’t experience pain, and the disease typically progresses very slowly.
Imagine how many people’s vision could be preserved if there was a way of showing them how their treatment program was working. Crucially, the system also requires little physician time. By 2025, 50 percent of the ophthalmic workforce will have retired – but the number of students entering residency is dropping off. Devices like ours will play an important role in tackling the increasing treatment burden, made worse by an aging population.
Our device may also have significant potential in terms of patient responsibility. Too many ophthalmologists are exposed to malpractice claims. Wouldn’t it be helpful if they could arm themselves with the evidence that it was, in fact, the patient who hasn’t complied with recommendations? That’s not to say we are going to act as a mediator between insurers and patients – rather, our data can empower patients to take their medication as prescribed, and everyone – physicians included – can benefit from the results.
Comments