Medo
AcquiredEntrepreneur First '17Singapore, SG · Founded 2017 · 4 known investors
Medo developed an FDA-cleared cloud platform that segmented ultrasound scans into three-dimensional anatomical views and returned findings in seconds. The interface was built so that operators without specialist sonography training could identify a range of common and urgent conditions. Exo acquired the company in 2022.
Also known as Medo AI · Medo.ai · MEDO.ai
Founders & leadership
Medo was founded in 2017 by Dornoosh Zonoobi.

Investors · 4
Company profile
researched Aug 2026Medo (Medo.ai) was a Singapore-based health technology company that combined artificial intelligence, 3D reconstruction and cloud computing to simplify the acquisition and interpretation of ultrasound images. Its stated aim was to make ultrasound function as "the stethoscope of the 21st century" by removing the requirement for a radiologist or trained sonographer to be present during a scan. The company cited World Health Organisation figures that nearly two-thirds of the world's population lacks access to basic medical imaging, and framed its product as a way to extend imaging to frontline workers such as family physicians, emergency room doctors and ambulance staff.
The core technique took 2D ultrasound images captured from different angles and reconstructed them into a 3D model of the anatomy, which the system then compared against thousands of prior cases to produce a suggested diagnosis. The company offered a web-based architecture in which clinicians could view and analyse images directly, including images acquired at remote healthcare centres. Medo's first and most advanced module addressed developmental dysplasia of the hip in infants. Additional use cases under development included detection and characterisation of thyroid nodules and early detection of masses and inflammation in the liver, for which data collection, labelling and validation had been completed as of the January 2020 interview.
During the Covid-19 pandemic Medo released Medo Lung, an algorithm trained to inspect lung ultrasounds for abnormalities indicating the severity of possible Covid-19 infection or pneumonia. The company presented this as an alternative to CT scanning, which involves radiation, patient transport out of the ward and machine sanitisation between uses; with Medo Lung a clinician could scan at the bedside with a handheld probe, upload to the cloud and receive results in seconds, supporting triage of patients by severity.
Founding story
Medo was co-founded by Dornoosh Zonoobi with radiologists Jeevesh Kapur and Jacob Jaremko; David Quail is also described in a 2021 interview as a co-founder. Zonoobi completed a PhD in medical machine learning at the National University of Singapore and accumulated over a decade of research experience in medical imaging. While working in a hospital radiology department in Edmonton, Canada, helping radiologists scan babies for hip dysplasia, she observed patients travelling thousands of kilometres, sometimes by helicopter, for a simple scan, and concluded that access to medical imaging was limited and centralised. That observation, combined with the portability and low cost of modern ultrasound probes relative to the expertise needed to interpret their output, led to the founding of the company.
Business model
Medo sold software rather than hardware, positioning its analysis layer as device-independent so it could work alongside existing and handheld ultrasound probes. Clinicians captured images at the point of care, uploaded them to Medo's cloud, and received analysis back, which allowed the company to serve remote healthcare centres without local imaging expertise or on-premise infrastructure.
As described in January 2020, the company expected that after FDA clearance the majority of its revenue would come from fee-per-study software licences, offered either as perpetual one-time licences or as software-as-a-service subscriptions.
Traction
Four multi-centre clinical studies confirmed the accuracy of the 3D surface models for hip dysplasia, drawing on more than 5,300 hip scans collected over five years in four countries; the technology detected developmental dysplasia of the hip more accurately than 2D scans with 75 percent fewer inconclusive cases, and a later study reported reliability with novice users. Two additional studies formed part of product verification and validation ahead of an FDA submission. A three-year efficacy and accuracy trial was completed in July 2017 and a University of Alberta trial with inexperienced operators began in 2018. Medo Lung was in regulatory approval processes and under trial with Canadian hospitals and clinics as of 2021.
Latest developments
A medical startup directory entry for MEDO.ai states that Exo acquired Medo AI. Prior public coverage documents Medo Lung being trialled in Canada and supported by the AWS Diagnostic Development Initiative.
▸Full profile — market position, technology, go-to-market, geography, history, risks & controversies
Market position
Medo was covered as one of a group of Singapore-based deep tech companies applying AI to medical imaging, alongside See-Mode Technologies, which addresses stroke risk prediction from ultrasound, CT and MRI. Medo was a portfolio company of SGInnovate, which was among its earliest investors.
Medo's stated differentiators were the reconstruction of 2D ultrasound sweeps into 3D anatomical models for comparison against large reference case sets, a device-independent web-based architecture, and a capture workflow described as being as simple as taking a panoramic photo on a phone. Company-cited studies reported more accurate detection of developmental hip dysplasia than 2D scans with 75 percent fewer inconclusive cases, and reported reliability when scans were performed by novice users.
Technology
The platform captured 2D ultrasound images from multiple angles and reconstructed them into 3D models of the anatomy, then compared those models against thousands of similar prior cases using machine learning to estimate the likelihood of disease. Processing ran in the cloud on a web-based, device-independent architecture, returning results within seconds and allowing remote viewing and analysis. Medo used Amazon Web Services infrastructure, including Amazon SageMaker for model training and Amazon Textract for extracting information from medical images, which allowed the team to iterate on algorithms without large server costs.
Go-to-market
Medo worked through clinical studies and hospital trial partners to validate its modules ahead of regulatory clearance, and pursued FDA clearance as the gateway to commercialisation. SGInnovate, one of its earliest investors, provided platforms for showcasing the technology to potential stakeholders and customers in the region and helped build connections and collaborations. The company positioned its product for smaller clinics, remote centres and pre-hospital settings such as ambulances that lack imaging expertise.
Frontline healthcare workers rather than imaging specialists: family physicians, emergency room doctors, ambulance staff, and clinicians at smaller clinics and remote healthcare centres lacking radiology expertise; hospitals seeking to triage and manage patient loads.
Geography
Headquartered in Singapore and described as a Singapore-based startup. Clinical work and trials were conducted in Canada, including at the University of Alberta and with Canadian hospitals and clinics for the lung product; hip dysplasia study data was collected across four countries. The company targeted US Food and Drug Administration clearance for market access.
History
The idea for Medo originated around 2015 while Zonoobi was working at a hospital in Edmonton, Canada, helping radiologists scan infants for hip dysplasia and observing patients travelling long distances for simple scans. She teamed with radiologists Jaremko and Kapur to build a cloud-based diagnostic aid. A three-year trial of the hip dysplasia technology was completed in July 2017, and in 2018 the company started a clinical trial at the University of Alberta assessing scans taken by users with no ultrasound experience. By early 2020 four multi-centre studies had been completed and two further verification and validation studies were in progress ahead of an FDA submission. During the Covid-19 pandemic Medo developed Medo Lung for lung ultrasound triage, working with hospitals and clinics in Canada and with support from AWS's Diagnostic Development Initiative. A medical startup directory records that Exo acquired Medo AI.
Risks & controversies
Company representatives identified access to large and diverse healthcare datasets as a core challenge, noting that legal ownership of data and the transfer of healthcare records to private entities are not clearly defined. They also cited the need to manage public and clinician opinion on AI in healthcare, the absence of clarity over accountability when AI systems err, and concern about AI-driven changes to clinical workflows and job roles. Commercialisation depended on regulatory clearance that was still pending at the time of the 2020 and 2021 sources.
Compiled by commissioned research from 8 cited public sources — announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed — not independently audited.
Timeline · 4
launches, deals, and filingsA medical startup directory listing for MEDO.ai notes that Exo acquired Medo AI.
Medo's lung ultrasound project was supported by Amazon Web Services' Diagnostic Development Initiative; the company used Amazon SageMaker to train AI models and Amazon Textract to extract information from medical images.
Medo introduced Medo Lung, an AI tool trained to inspect lung ultrasounds for abnormalities indicating the severity of possible Covid-19 infection or pneumonia; scans are uploaded to a cloud-based system that returns a result within seconds. At the time of the interview the tool was in the process of obtaining regulatory approvals and was being trialled with hospitals and clinics in Canada.
Medo began a clinical trial at the University of Alberta, Canada, evaluating ultrasound scans taken by people with no experience using the device, following a three-year trial completed in July 2017 that assessed the technology's efficacy and accuracy at detecting hip dysplasia.
Dated company events from announcements, filings, and press; legal rows summarize public dockets and regulator releases.
▸Research sources · 8
primary sources listed
- MeDo - Build full-stack Apps With No-Code AI Platformmedo.dev · web
8 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Medo do?
- Singapore health-tech company using AI, 3D reconstruction and cloud computing to make ultrasound interpretable by non-specialists.
- Who founded Medo?
- Medo was founded by Dornoosh Zonoobi in 2017.
- Who are Medo's investors?
- Medo's investors include Edmonton Unlimited, Elev8.vc, Entrepreneur First, Qualgro.
- Where is Medo headquartered?
- Medo is headquartered in Singapore, SG.