Recent investigations
into US, UK and Australian court cases have revealed that too many innocent people were being wrongly sentenced through gaps in understanding.
The three year research project aims to improve forensic evidence, and has the potential to transform ballistics investigations in Australia and globally.
The project will use machine learning to create a digital 3D model of the human anatomy, including entry and exit wounds, allowing investigators to record the trajectory of the projectile through the body, identifying and localising projectile fragments.
In an interview with Innovation Intelligence, Associate Professor Richard Bassed explained how they planned to combine machine learning in their approach.
“The reason we could do this is our database where people have died of all courses of death, all ages, all sort of aspects: weight, size, height. A database of 80,000, that we use as a training set for a number of algorithms and diagnostics of trauma and disease.”
“Ballistics in forensic medicine has traditionally involved fairly basic analytical techniques, which have not changed for a century,” he said.
Pathologists look for entry and exit wounds to see where the bullet is. Sometimes pathologists need to search for it manually, which can be an invasive process, as the bullet hasn’t gone through the exit wound or may have moved.
In the long-term, the technology may be able to help investigators determine the calibre of bullet, and distance the shot was fired from. This will allow investigators to determine the type of gun used, and if the wounds were self-inflicted or resulting from attempted homicide.
Bassed said that they would also be able to determine the trajectory path of other kinds of trauma wounds.
“Once you’ve got the data you can develop algorithms for just about anything- any sort of foreign object sleight on the body you can access. It could be used for motor vehicle accidents where people can be killed or impaled by various parts of the car.”
Chris Bain, Professor of Practice in Digital Health in the Faculty of Information Technology and Monash University’s Lead for Digital Health, said the project was just one example of how artificial intelligence and data science were transforming the digital health and forensic spaces.
“This approach is much more scientific and rigorous than the way this procedure is currently performed.”
“This technology has the potential to scan and analyse the body, as opposed to the body being dissected. The technology could streamline workload and time efficiencies; and address any cultural sensitivities that may arise.”
With further testing and upscaling, this technology could reduce the need for post-mortems for shooting victims allowing for cultural sensitivity. This may include families that don’t want their loved ones being subjected to an autopsy or religious burial rituals that call for the deceased to be buried at a certain time after death.
Information technology and biomedical research company Leidos has contributed $150,000 to the project which was made possible through the Monash Institute of Medical Engineering (MIME).
The initiative comes from Monash Data Futures, an institute focused on using AI and machine learning to drive change in health science, sustainability, governance and policy.