Advanced Imaging and Modeling Lab

The Advanced Imaging and Modeling (AIM) Lab has transformed pre-surgical planning through 3D modeling. Initially, surgeons would hold a 3D printed heart in their hand the night before they saw the same heart in the operating room (OR); now surgeons are using even more advanced tools while reviewing 3D digital versions of hearts and other anatomies in virtual and augmented reality. The OSF HealthCare Children’s Hospital of Illinois has uniquely seen the impact of these advanced digital twin technologies in pre-surgical applications and is redefining the standard of care for pediatric surgical care.

The new standard of care at OSF Healthcare

Impact of 3D modeling

Since 2013, the AIM lab has committed to providing OSF Children’s Hospital surgeons with digital twin anatomy for pre-surgical analysis.  Through this commitment, the AIM lab has amassed the largest experience in pre-surgical digital twin utilization nationally with over 500 cases of digital twin transformation from congenital heart disease, orthopedics, neurosurgery and surgical oncology.  This volume of success is related to two distinct factors; the access to cutting edge technology and commitment from the AIM lab and innovative surgical leadership at OSF Children’s Hospital.  The surgeons at OSF Children’s Hospital are the main reason this technology spread from the lab to the OR.  Once a surgeon experienced a unique pre-surgical insight gained from a digital twin, they immediately re-drew the line of standard of care to include this technology.  This clinical care advocacy is unique to the health care resources available here in Peoria, IL.  Between OSF Healthcare, Jump Simulation, University of Illinois College of Medicine Peoria and University of Illinois Urbana-Champaign, this community benefits from the fertile ground needed to grow a world-wide leadership position in state-of-the-art digital twin pre-surgical planning. 

This success at OSF Children’s Hospital is now being expanded system-wide at OSF for an ever-expanding list of complex surgical specialties.

Machining the future

The expertise needed to provide this service for a single patient, let alone an entire hospital or hospital system, remains a tremendous barrier to broader adoption.  As surgeon after surgeon returned to the AIM lab stating they need this technology, the AIM lab committed research efforts to solving the barriers.  In collaboration with the University of Illinois College of Engineering, the AIM lab turned its attention toward the complex problem of automating the conversion of standard CT and MRIs into 3D digital twins of patient specific anatomies.  In 2013, each model was generated from a CT or MRI by a radiologist or cardiologist painting on the radiograph, slice by slice, each patient specific organ.  When these painted images were stacked, an accurate 3D model emerged which was then printed or viewed in VR. 

Since the AIM lab’s inception, this expertise reliant step has been the major target of automation.  Computer vision, or the ability of a computer to replicate a radiologist’s visual understanding of a medical image, began to mature to a level of plausibility around this time.  By 2018, the first machine learning, or computer vision, or artificial intelligence algorithm, was developed by the AIM lab to begin to solve this problem.  Over the past several years, this technology has advanced significantly.  The first AIM lab machine learning solution took 5 years, the second solution took 2 years, and the most recent solution occurred over this past summer by an intern.  Today, the AIM lab deploys machine learning tools to automate the creation of patient specific digital twins at an exponential rate. 

If automated segmentation is step one of improved radiological interpretation through digital twin creation, the AIM lab is already working on step two of automated analysis and diagnosis.  The AIM lab has published work demonstrating that a machine can replicate the diagnosis of a radiologist based on patient specific 3D model analysis.  As more and more 3D and 4D digital twin representations of anatomy are created, the opportunity for new automated diagnostic technologies emerge and the AIM lab is already working on the next advancement; the machine learned radiologist.

If you are interested in learning more about the Advance Imaging and Modeling Lab, participating in certain aspects of the journey or if you want to invest or collaborate, contact us today.

Leading the Way