The world of orthopaedic knee surgery has taken a giant leap forward with AI. With improved precision and faster procedures, Alexandra Hospital’s patented AI algorithm is set to change the future of knee replacements.
Alexandra Hospital in Singapore has developed a novel AI algorithm that is revolutionising robotic total knee replacement surgery, with the potential to greatly improve patient outcomes.
This technology, which has secured both local and international patents, is recognised to a unique computational algorithm designed to improve the accuracy and efficiency of robotic total knee replacements.
The Challenge in Knee Replacement Surgery
Total knee replacement surgery is one of the most common orthopaedic procedures, particularly among ageing populations. In Singapore alone, the prevalence of knee replacement surgeries has surged, increasing 2.7 times from 1999 to 2019. Patients suffering from severe knee pain, often due to osteoarthritis, rely on these procedures to restore mobility and quality of life.
However, even with the rise of robotic-assisted knee replacements, challenges remain. Traditionally, surgeons manually adjust the positioning of implants during surgery – a complex process requiring consideration of eight degrees of movement within the knee joint. These adjustments influence one another, making it difficult to achieve optimal implant positioning.
The intra-operative planning process can take up to 15 minutes – taking up precious time in the operating theatre. Surgeons often have to make do with a solution that is close to ideal, rather than the ideal, because of time constraints in the operating theatre.
This variability can lead to suboptimal outcomes. Implants that are not placed most optimally are at greater risk of wear and tear and may require revision surgery down the road.
Precision in Knee Surgery Powered By AI
Developed by Dr Glen Liau Zi Qiang, an orthopaedic surgery consultant at Alexandra Hospital, in collaboration with researchers Dr Matthew Ng Song Peng and Mr Ryan Loke Wai Keong, the novel AI algorithm tackles these challenges head-on.
This algorithm is integrated into the cutting robot’s software. With AI, it can compute thousands of permutations to identify the optimal implant position in less than 0.1 seconds. The algorithm takes into account the patient’s unique anatomy, as well as the surgeon’s targets for implant positioning.
Study Design & Results
A prospective study was done, involving 67 patients who underwent robotic total knee replacement surgery from November 2021 to December 2023. Of these patients, 25 had the algorithm utilised during their surgery, while 42 did not. It was shown that:
- High Accuracy: The algorithm achieves 92% accuracy in implant placement, compared to 52% in the manually planned group.
- Computational Efficiency: The algorithm can solve for thousands of solutions in just 0.1 seconds. It then ranks optimal solutions, shortlisting the top solutions for surgeons to trial intra-operatively. This is in contrast with the manual planning, which may test for only 1 to 2 permutations.
- Faster Surgery Times: Procedures using the AI system are completed 48% faster than manual ones – from an average duration of an hour plus to under 40 minutes.
Beyond improved patient outcomes, the increased efficiency also means that the surgeon is cognitively and physically less fatigued, allowing for quicker procedures. This enables each surgeon to perform more operations, benefiting more patients.
As a result, the overall productivity of the healthcare provider, hospital, and healthcare system is elevated.
Improved Patient Outcomes
The increased precision of AI-driven robotic knee surgery has the potential to significantly improve patient outcomes.
- More accurate implant placement could lead to better post-operative results and longer lasting implants.
- As the AI algorithm reduces surgical time, patients may experience reduced blood loss and shorter anaesthesia duration, reducing surgical risk.
- The improved accuracy reduces the risk of post-operative complications like loosening, wear, and instability.
Ultimately, patients may experience higher satisfaction with their knee replacements due to better alignment and fewer complications.
Mdm Rita had suffered from knee pain for years, relying on painkillers and injections to cope. She had done her first AI-powered knee surgery just a year ago.
Recounting her first surgery, she shared. “Within a week I was able to walk…the recovery was very fast, I was happy with it.”
Encouraged by the results, she went for the second surgery a year later. Just two weeks after the second AI-powered knee surgery, she was walking unassisted at the media session.
“It was good and no issue so far so.” She said with a smile. “My big thanks to Dr Glen and his [team].”
Since its implementation in August 2023, approximately 200 patients have undergone knee replacement surgery using this AI implant placement method. Studies are ongoing to validate its long-term impact
AI in the Future of Surgery Practices
The development of this AI algorithm has the potential to set new standards in robotic-assisted orthopaedic surgery, ultimately benefiting both patients and surgeons worldwide. The technology is not limited to one robotic system. The team is in talks to develop it for all major systems and implement it in a broader scale.
“I am passionate about incorporating innovative solutions to enhance my patient outcomes.”
said Dr Glen Liau who practices and operates at both AH and NUH.
“It brings me great joy and satisfaction to see my patients enjoy a significantly improved quality of life following surgery.” he added.
While the long-term clinical outcomes are still being studied, this innovation is a step towards more precise, efficient, and effective knee replacement surgeries. This is an exciting development, and as AI continues to transform medical practices, it holds great promise for significantly enhancing patient care.
