Advancing Healthcare through Open-Source and Data Sharing in Medical Device Development
Openwater is enabling collaboration via open source development and sharing of safety data to reduce cost and speed rapid development and regulatory approval of new technologies. We have made all of our research, patents, architectures, design and software open-source.
​
Why we’re doing this: The landscape of medical device development is a tough one: each developer usually creates their own hardware and software and manufacturing supply chain and then this typically goes to just one medical indication.
​
We see a different approach: manufacturers, developers, and researchers stand to gain substantial benefits from the proactive sharing of architectures and safety data, while achieving higher production volume with our approach should enable us to make high quality semiconductor components for use by all. This collaborative approach both fosters a culture of transparency and innovation and points toward advantages in regulatory compliance, cost savings, and the ability to leverage safety data across multiple clinical trials.​​​​
How: Our business model may speeds regulatory approvals in addition to consumer electronics pricing at speed. Today's average novel PMA therapeutic device capitalized cost is 13 years and $658M in 2024 dollars. This is a comprehensive study of PMA therapeutic devices to achieve FDA approval in the table below. 85% of that $658M is in the device development labeled "Non-clinical: device development" in this table
​​​
The remaining 15% of the average capitalized cost of an FDA approved novel PMA therapeutic medical devices is in acquisition of safety and efficacy data. On average 607 patients are needed for trials that that span 8.6 years. Our devices enable reduced the number of patients and the elapsed time needed:
​
1) The average number of patients estimated for therapeutic ultrasound clinical approval is 50-150 patients, This reduces cost for trials not just in numbers of patients needed, but also in the elapsed time and their associated capitalized costs . For example:
​
-
Trials leading to the FDA approval of treatment of essential tremor in 2016 involved 76 patients
-
Sonoblate's FDA approval for prostrate cancer treatment was based on clinical data from 117 patients in a multicenter trials. Further discussion here
-
ExAblate acheived FDA approved for the non-invasive treatment of uterine fibroids. The approval was supported by a pivotal study involving 109 patients, and later FDA approval for the palliation of pain in patients with bone metastases via a pivotal trials of 148 patients
-
Exogen achieved FDA approval for the treatment of non-union bone fractures in clinical trials involving 74 patients.
​​
2) Shared safety data pan-disease (like pan-cancer or pan-mental disease): the literature suggests a further reduction in needed patients and the equivalent reduction in elapsed time required (and thus in capitalized costs). Several studies make these estimates often citing potential sample-size reductions ranging from 20%-50% when using shared control arms, master protocols, or pooled safety data. However concrete numbers tied to specific studies are scarce in the public domain. We cite 3 such studies below.
​
-
​Saville BR, Berry SM. “Efficient Adaptive Clinical Trial Designs.” In: Chi G-YH, Jeong J-H (Eds). Medical Statistics: A Guide to Data Analysis and Critical Appraisal. Basel: S. Karger AG; 2021, p. 236–252.​
-
Key Point: Provides an overview of Bayesian adaptive designs for platform trials. While not presenting a single “percentage” table, it includes modeling scenarios showing that using a shared control arm and early stopping rules can lead to significant reductions in total sample size—commonly cited figures are in the 20–50% range.
-
​
-
Saville BR, Berry SM. “Platform Trials in Drug Development: Umbrella Trials and Basket Trials.”​​
-
Source: A tutorial presented at clinical research conferences (various years, slides often available through Berry Consultants’ website or workshop proceedings).
-
Key Point: Illustrates how multi-arm, multi-stage platform trials with a single (shared) control can reduce the number of control-arm participants required for multiple treatment arms. In example calculations, they note a 30–50% reduction compared to running separate two-arm trials.
-
​
-
Berry SM, Connor JT, Lewis RJ. “The platform trial: an efficient strategy for evaluating multiple treatments.” JAMA. 2015; 313(16):1619–20.
-
Key Point: Introduces the concept of platform trials and discusses the potential for fewer patients overall by sharing infrastructure and controls. While the editorial doesn’t quote exact percentages, it’s a concise overview that underpins subsequent presentations where the “30–50%” figure is often cited.
-
​Summary of these estimates below:
Our open-source and reconfigurability is fundamental to our business model. Our highly reconfigurable platform has life-saving potential to deliver therapies for cancers, mental diseases, cardiovascular diseases and much more sharing across a sub-field of diseases to accelerate approvals - via the same Openwater open-source platform.
This approach may massively lower the cost and time of development for each therapy, beyond the ~100x we have already taken out of device cost itself. Our goal: save more lives more quickly - regulatory approvals with sub-$10M spending and sub-3 year timelines by our customers using our platform.
​
Note: this is a hypothesis based on the above data we seek feedback from all to refine and improve our estimates.