Interview with Roxana Rusu-Both, Co-Founder @ LEDD

How would you summarize what your startup does in a brief way?

LEDD - Lung Ecography Differential Diagnosis - is an image processing and AI based lung ultrasound (US) differential diagnosis and monitoring application that aids physicians in the lung cancer (LC) and interstitial lung disease (ILD) diagnosis process. It is able to identify the lung lesion and give a characterization of the lung lesion, a malignancy probability and a clinical medical criteria evaluation. We developed medical differentiation criteria based on pathologic changes visualized on US which were evaluated with own algorithms as image analysis indexes. AI was used next for diagnostic indication. The main benefits are: 1) Non-invasive, radiation free; 2) Widely accessible; 3) Safe for patient follow-up with a higher frequency; 4) Enables patient stratification; 5) could reduce healthcare cost; 6) support clinical and treatment decisions; 7) could improve life expectancy; 6) Evaluation of non-transportable patients. The MVP was tested on 113 real cases with promising results (accuracy >97%). The applicability is high, both for diagnosis in primary care, internal medicine and for patient monitoring after medical treatment is administrated in internal medicine and oncology.

What is unique about your product?

A complete examination of the lung lesion by ultrasound (US) can provide information on its benign or malignant character and also can be used to evaluate patients with lung interstitial damage. The LEDD approach of using AI/ML and thoracic ultrasonography for lung cancer diagnosis and follow-up investigations and interstitial lung disease evaluation is a completely new approach providing higher quality care while minimizing total treatment cost. While new and original, the LEDD approach is based on solid medical research and experimentally proven by validations tests on real cases. The main benefits are: 1) Non-invasive, radiation free; 2) Widely accessible; 3) Safe for patient surveillance; 4) Enables patient stratification; 5) could reduce healthcare costs; 6) support clinical and treatment decisions; 7) could improve life expectancy; 6) Evaluation of non-transportable patients.

In which stage of development is it?

Currently, the expertise of the team members and the preliminary knowledge situate the solution in the TRL4 (IML3) category. A small-scale, MVP prototype integrated with complementing subsystems at laboratory level was developed. The validation of the new technology was performed on 113 real cases enrolled. Prototype shows stable performance obtaining: Precision>96%, Specificity >90% and Accuracy>97%.

Tell us a bit about the team. How important is the diversity of the team for you?

AIMED Soft Solution team is multidisciplinary composed of 6 specialists from medical & software development field. Our team evolved since 2017 and now includes medical clinicians and experts in applying artificial intelligence algorithms. Our deep understanding of clinicians allows us to efficiently harnesses the power of artificial intelligence to deliver accurate and robust clinical decision support systems based on standard clinical ultrasound imaging that can fit into clinical workflow. Assoc. Prof. Eng. Rusu-Both Roxana, PhD –CO-Founder, CEO&CTO has a wide experience in coordinating research projects. Prof. Dr. Chira Romeo Ioan, PhD –CO-Founder, CMO is a senior physician, ultrasonography expert, researcher and Head of 1st Medical Clinic, SCJU, MSc. Rusu Tudor-Andrei - COO has 12 years’ experience in software & product development with a degree in medical physics.

What difficulties have you experienced so far?   

The main challenges consist in fund raising, business development, market analysis and market penetration. Currently, we are not operating under any Regulatory Clearance, our solution being in the TRL4 (IML3) maturity level. The next step is to obtain a CE mark in accordance with the new Medical Devices Regulation (MDR). For this we expect it to be necessary to perform a multicenter clinical study. The certification represents our biggest challenge for which we need expert help to guide us through the necessary steps.

What went better than you expected?   

The development of solution and the MVP went better than expected.

In the most favorable scenario, how does your startup look like in the next 2 years? What might not work as planned?

In the next 2 years we expect to have a validated solution, CE certification and market penetration at least in one country. The time to market could increase due to some barriers like: funding, administration, no. of stakeholder: hospital manager, head of specialty and overall budget owner, high no. of projects for country/region or even hospital centric strategies.

What do you require at the moment to accelerate your development?

Our needs could be divided in multiple categories but are mainly represented by because we can used it to sustain the clinical trial, CE certification, the development of the market ready product and also expertise (people) in business, business plan development, product certification, marketing (market research) and commercialization. Up to now we started working on all of these topics and activities but we still need help and guidance from more experienced mentors.

How was the concept born?

Usually, medtech projects have to start from a need identified by medical specialists. This was our case. I found this topic very interesting with high applicability. So, we applied for national funding and ultimately, we started collaborating in 2017 on a national research project. During this project we developed the proof of concept. After that the team extended working on 2 RIS Innovation call projects funded by EIT Health in 2019, 2020 leading to the MVP of our solution. In 2021 participated in the Innostars Awards program and we were in the top 10 finalists.

Something you haven't shared yet about the startup

Because all of us are from academic environment we all have teaching responsibilities and our schedule is quite different even if we allocate time for the product development. This is why we have “night meetings” not daily meetings because they are usually scheduled from 10 pm.