DoseMe eyes machine learning to boost analytics platform

Australian healthcare startup turns to Azure to help drive its SaaS offerings

Queensland-based startup DoseMe is looking at leveraging cloud-based machine learning services courtesy of Microsoft Azure to bolster its analytics platform, DoseMe Crunch.

DoseMe’s primary software-as-a-service offering is designed to make it easier for clinicians to calculate the correct dose of a medication for an individual patient, said founder Robert McLeay.

According to the company, it is the only software cleared by regulators in Australia and Europe to deliver personalised dosing recommendations to clinicians.

“We know that we’re all different; there’s no such thing as average,” McLeay said. “When it comes to dosing a drug it’s not how we look on the outside that counts — it’s how our metabolisms work on the inside.”

DoseMe takes information about an individual including data such as height and weight, any doses of a drug that he or she has previously been given, and any pathology results and plugs that into a mathematical model that can calculate how they will respond to a particular dose of a medication.

“Once you can measure the effect that a dose has on you, then you can calculate a dose that's designed to deliver the effect or the amount of response that a clinician would like to have.”

“Our target markets really are pharmacists, doctors and hospitals,” McLeay said. “They’re the ones that need to calculate the dose of drugs for you to take, and currently some of the ways that they do that are slow, manual and laborious, not to mention are simply not personalised.”

St Vincent’s Hospital in Sydney is one of the service’s major customers, and DoseMe is also being used in a number of regional hospitals such as Barwon Health in Victoria, he said.

The company's complementary analytics platform DoseMe Crunch can be used to aggregate, de-identify, cleanse and analyse an organisation’s clinical data using the same technology that underpins the company’s main service. DoseMe Crunch offers opportunities for organisations to benchmark themselves and leverage predictive analytics features.

The service can help a hospital develop a detailed model of how patients have previously responded to a drug in order to find areas for improvement.

DoseMe runs entirely on Microsoft’s Azure cloud computing service out of local data centres, with cloud proving to be a natural fit for the application, McLeay said. (DoseMe was originally hosted on Amazon’s public cloud.)

“Health care data security and sovereignty within a given country is absolutely critical; it's probably one of the most frequently asked questions by our customers, and Azure is a fantastic partner to work with for that very reason,” the DoseMe founder said.

“If you were to do this the traditional old way, when we switch on a new site we'd have to buy more physical servers and stick them somewhere. We need to get our facility audited for the health care requirements that we need — or we could simply use a cloud provider like Microsoft Azure and go, ‘You know what? We need to switch on some additional infrastructure.’”

“The partnership and marketplace model that Azure has gone with makes it really easy for us as we grow and need a new service,” he added. “For instance we’re currently in the process of migrating all of our transactional emails over to Azure’s platform.”

Join the Computerworld newsletter!

Error: Please check your email address.

Tags dataazurehealthCase studiesCloudHealthcarebig datamachine learningcloud computingCase Studymicrosoft azure

More about Microsoft

Show Comments

Market Place