The Coach

With the digital diary in hand, and Checkpoint as a user-friendly analysis tool for the diary data, it was time to move on to the therapy part we were all waiting for — and I already had the solution. Fortunately, there is something called urotherapy: it includes non-pharmacological and non-surgical methods to treat LUTD. It consists of behavioral changes to improve the urination schedule, adjust fluid intake, take the proper position while peeing, or make some specific effort to completely empty the bladder. These recommendations also include improving bowel habits with the proper position for emptying the bowel, and regular toilet visits after meals. So if the diary can deliver data on the cause of LUTS, then a smart and carefully designed algorithm could deliver personalized remote urotherapy — that was the plan. An excellent part about urotherapy is that it is regularly investigated in clinical studies, which have shown similar outcomes to pharmacotherapy. Urotherapy is a first-line recommendation in every urological guideline for treating LUTS, both for adults and for children. What's more interesting, there were trained teams in some hospitals, primarily in the Netherlands, that already did remote urotherapy with on-site initiation and later telephone calls. So if delivered properly, our app could be a game changer in digital urotherapy.

In practice, some cases — such as extraordinary daytime frequency or post-void dribble — can be treated with urotherapy alone, with complete success in a short time. Other cases, such as overactive bladder, stress incontinence, or giggle incontinence, may need longer adherence and have a smaller success rate. Bedwetting is probably the hardest to resolve with urotherapy alone, but following a prescribed plan may help reach success faster when coupled with other tools, such as a bedwetting alarm. So the idea of developing a remote, personalized urotherapy plan based on diary data was sound — this could provide users of the BeDRY app with the tool they actually wanted in the first place. I used “bladder training” as an umbrella term for both standard and specific urotherapy when writing the grant proposal. Bladder training is sometimes used to refer to a specific part of urotherapy used in extraordinary daytime frequency — when the patient goes to the bathroom often, but without incontinence if urination is postponed, and these patients do not have nighttime symptoms such as bedwetting. Bladder training in that case means postponing urination by 15 to 30 minutes to train the bladder and reduce urgency sensation, and this approach is effective in that condition. However, if a patient has urge incontinence, this is primarily overactive bladder (if urinary tract infection is excluded). So, as you can see, the right medical history — and data on daytime and nighttime peeing habits and accidents — can make a huge difference in providing rational urotherapy: prescribing the optimal treatment for the condition. The term “bladder training” seemed easier to understand for people without medical education when talking about urotherapy, so I kept it.

One year before all the work on the Checkpoint system, I had applied for an EU grant for a proof-of-concept study. I presented a project with the goal of developing a digital model of personalized urotherapy and testing it at Technology Readiness Level 4 — meaning I would prove its validity in experimental settings. It was not a fully non-refundable grant, and I had to participate with my own funds. The project grant sum was large at the moment — just over 50,000 euros — and I didn't have 30% of the money I needed for participation. I had already spent the money from selling my car to fund the development of the MVP, and I was now officially in the FFF region of financing — friends, family, and fools — as I used to joke back then. Honestly, it wasn't that funny at all. I had already invested in the web app, in startup capital, in working hours to fix bugs, and had published a mobile app on Google Play. I barely had money for the monthly bills to keep the working app online. This was for sure the lowest point in the life of my app: I had already stretched too thin, and my financial commitments had exceeded the safe zone. It was the point where I was sure just a moment separated me from failure, unless some miracle happened.

If I could find the money to ensure participation in this grant, and if I won the grant, it would be a huge thing. My app was in its earliest phase, only one year old. No one actually saw any value in it — that includes both my friends and family — and I had run out of fools.

At that time, my late mother was a terminal cancer patient, and I spent at least one hour visiting her every day. I would usually bring one of my children with me, so they could have memories of her when she passes away. I told her I would ask my father for a loan, and I needed “only” 10,000 euros. My mother had two prominent characteristics: she gave advice even when you never asked for any, and she was a problem solver. When I visited her the next day, she told me not to ask my father — she would give me the money I needed. “Your father will be alone when I die, and will need his savings to live the rest of his life without depending financially on anyone,” she told me. In the following months, on what would be her last trip to her home town, she went to the bank and later handed me a hefty envelope. She kept her promise, and she was the first investor in BeDRY. She believed in me and wanted to see me succeed and follow my dreams. When she passed away, a couple of days before her 74th birthday, I got an email from the funding agency that my grant proposal was accepted. I was sure my mother was watching me from heaven and pushing circumstances a little in my favour. Love you, Mom.

The proof-of-concept project, which I named BeDRY Diary & Bladder Training, started in January 2025. By then I had already prepared the whole bladder training concept as a Word document, with technical data summarized in Excel sheets, and mock-ups in PowerPoint slides. All of this was needed to present the idea to Kristijan, my lead dev. The most useful was the Excel sheet — there I had brainstormed nine interventional categories with different urotherapy recommendations. There was a table explaining how the algorithm should pick the correct intervention group based on recorded diary data. There was also a list of separate urotherapy tips. Every intervention category had its own list, which I could preview in a separate sheet to check the output of the concept. In the outpatient clinic, with a child sitting in front of me, I would open the file and type in the age, the gender, the weight. Excel would calculate how often this child should be peeing and how much they should be drinking, and I would walk through the appropriate sheet to choose interventions. It was a punched-tape calculator — my own ENIAC for urotherapy, only digital. I knew which interventional category each patient should belong to, so checking the appropriate interventional sheet, I could correct and update my urotherapy items and how they clustered between intervention groups. After twenty patients, patterns started to repeat, and I sensed we needed to enter the next phase of development.

Meanwhile, Kristijan developed an even better tool, where I could add, edit, and assign urotherapy items to intervention groups, together with an algorithm builder and an output that presented the simple version of Checkpoint, the urotherapy tips, and the Diary report — it was a great work. I could review diaries that parents of my patients had agreed to share with me, and see the results of our digital tool immediately. That was the environment in which to test our concept.

Since the backend of the app was already finished, I had to work on the frontend — the part that users see and interact with. Another designer, Rea, agreed to animate our badges and create animations for urotherapy tips. We presented urotherapy tips in three categories: primary, secondary, and video tips. We pushed for this design after some user cases showed that the therapy report could overwhelm families with the amount of information provided, so I wanted to group recommendations into different sections based on their importance and the way they were presented. We showed only one to three primary tips; secondary tips appeared on one line and rotated randomly. Video tips opened an animation window where a title, description, and video were visible for each step of the recommendation. In practice it looked like this: for a girl with post-void dribble (caused by a small amount of urine entering the vagina during urination and then dribbling out when she stands up), the animation showed three steps of the correct sitting position to prevent it. I am grateful for these animations, because I believe Rea captured my exact thoughts on how to create them.