BRISBANE, Australia – Australian researchers are using artificial intelligence (AI) to diagnose medical conditions ranging from breast cancer to post-traumatic stress disorder (PTSD) more quickly than ever before.

Brisbane's Translational Research Institute (TRI) is working with Siemens Heathineers at Draper Laboratories using magnetic resonance spectroscopy to learn more about the chemical content of tissues and organs, providing a deeper understanding and earlier detection of conditions like PTSD, TRI CEO Carolyn Mountford said during the recent AusBiotech conference.

Using AI, clinicians will be able to identify the changes in a patient's brain chemistry associated with PTSD.

Mountford said the newly developed scanning technologies provide a better platform for diagnosing, monitoring and treating cancer, trauma, immunotherapy, chronic pain and other disorders and diseases.

"This will be a game-changer, as we will soon have the technology available to us to have a non-invasive test for PTSD to determine if soldiers and emergency workers are prone to the disorder and if so, can be rested or not immediately deployed again," she said.

"The idea behind machine learning is that it's a method that should generalize and be able to make decisions or make estimates or classifications based on data it hasn't seen before," said John Irvine, chief scientist for data analytics at Draper.

Various steps go into the machine learning process such as developing rules or algorithms, and validation of new data is the critical element in that process, he said.

For the TRI study, factors include PTSD and blast exposure to develop the algorithm, and then classifications are generated based on data collected from patients. Those classifications can be compared to inform clinicians.

Basically, the program builds on existing modalities and analyzes data normally obtained by patients to provide an automated analysis.

The U.S. Department of Defense and Australian military are working with Mountford's team to develop the new magnetic resonance spectroscopy (MRS) in vivo to diagnose changes to brain chemistry associated with brain injury and PTSD.

The team is identifying biomarkers that distinguish PTSD and mild traumatic brain injury using advanced MRS.

Mountford said that people have always evaluated the worst-case scenarios such as full-blown PTSD, and her goal is to be able to diagnose someone who has acute anxiety and "change their circumstances before they creep into PTSD."

"Once someone goes into that state, it's hard to get them out of it, but by identifying acute pain and acute anxiety, we can catch them before they commit to PTSD," she said.

By using a scanner at deployment sites, soldiers could be scanned before they are deployed, and then scanned again when they return, she said.

She stressed the need to provide clinicians with information in a way they can interpret visually. She said the move into using clinician-assisted diagnostics will be slow and defined, "and everyone will be looking for mistakes, and there will be mistakes," she said.

MRS also offers the ability to assess women at high risk of breast cancer in whom very early tissue changes can be identified, Mountford said. The technology can identify metabolic deregulations in breast tissue that precede tumor growth, detecting the cellular changes leading to the disease years before the cancer emerges.

Using AI to sequence the human genome

Another AI application spun out of Australia's QIMR Berghofer Medical Research Institute is a human sequencing project aimed at developing clinical assays for cancer patients and diagnostics labs, said Nic Waddell, head of QIMR Berghofer's Medical Genomics group.

She co-founded startup Genomiqa, which will offer hospitals, clinicians and pharma companies analysis of data from whole genome sequencing. She said the group can sequence the whole human genome in a few days.

Co-founder John Pearson, who leads the genome informatics group at QIMR, created the software for medical researchers and developed analysis pipelines for whole genome, exome and panel sequencing.

The group is working with big data company Max Kelsen to better predict cancer treatment outcomes harnessing AI and genomics.

Despite attempts to improve patient outcomes, median patient survival rates for some cancers remains low. And, although immunotherapy shows great promise in improving patient survival for some cancer types, including melanoma and lung cancer, it remains out of reach for many patients due to the prohibitive cost.

The Max Kelsen and Genomiqa collaboration is focused on identifying reliable markers to develop tests that predict which patients will benefit most from immunotherapy prior to treatment.

Funded by the Australian government, the immunotherapy outcome prediction (IOP) project is integrating AI and whole-genome sequencing into cancer research and clinical practice.

The key to predicting patient treatment outcomes lies in finding and interpreting the patterns and genes of significance in the genomes of patients who have responded best to previous treatments.

Brent Maxwell, a solutions architect at Amazon Web Services (AWS) Australia, said that Amazon is offering AI services to biotech companies that allow researchers to perform machine learning to calculate whatever they're interested in.

Amazon's Biotech Blueprint Quick Start can build a preclinical, cloud-based infrastructure on the AWS Cloud that is basically a "biotech-in-a-box," he said. It's based on industry best practices and can be used to build a biotech's infrastructure and configure it for identity management, access control, encryption key management, network configuration, logging, alarms, partitioned environments and built-in compliance auditing.

The AWS Benchling informatics platform allows life sciences companies to manage the workflow of their R&D processes and includes molecular biology tools, lab notebooks and sample tracking.