Introduction
Would you believe me if I said that because of the assistance of Artificial Intelligence (AI) in paediatric surgery, one-fifth of young patients have spent less time in recovery? A study funded by The Urology Foundation and led by scientists at the UCL and the University of Sheffield proved that this was the case. Not only that, but the incorporation of AI robotics in surgery procedures has halved the number of readmissions a patient might have.1 As we see the use of AI expand all around us, the medical world is a prime example of how AI assistance can really help make a difference - from performing surgeries with impressive precision to helping doctors diagnose diseases faster. With the goal of ensuring things run smoothly and efficiently, while constantly improving itself, the assistance of AI is a powerful co-pilot for a wide range of medical professionals to have in their practice.
This article will dive a little deeper into the history of AI and why it can play an important role in the medical field, as well as outline a few AI-based innovations that are currently being tested or used in the real world. With all that said, the use of AI implementation is still a very new development in many areas of discipline and therefore still brings various challenges to how and when it is utilised. Several ethical considerations and challenges are discussed below, as well as examining what the future landscape of paediatric surgery with AI could look like.
Understanding AI and its role in paediatric surgery
Artificial intelligence is any piece of technology that enables machines (including computers) to think like us and solve problems using methods and strategies that we would use.2 Think of AI as a chef’s assistant; a skilled assistant can help a master chef by handling repetitive tasks (such as washing, cutting, and prepping vegetables), taking stock of what ingredients are in the pantry, and ensuring everything runs smoothly in the kitchen. Arguably the biggest benefit of using AI is its ability to automate common tasks, analyse large sizes of data, and summarise useful insights. Because of this, our focus can pivot towards more complex and creative challenges. The two most well-known ways AI is being used in medicine are by supporting clinical decisions and aiding in medical image analysis. For example, since AI has the ability to learn and store a patient’s preferences (e.g., medication, treatments, check-up availability) it can provide personalised recommendations any time a patient wants them. Have you ever felt frustrated that you had to repeat your medical history to different individuals? Precision medicine could become a lot easier with AI virtual assistants that answer your questions based on your unique health history. Similarly, using AI to read medical imaging can help free up a medical professional’s time; AI can handle a huge number of medical images in a short amount of time and present the important and relevant parts to doctors.3
As we will see below, AI is particularly beneficial to paediatric surgery because it can enhance clinical diagnostics, provide real-time guidance in (robotic) surgery, predict surgical outcomes, and personalise rehabilitation plans. These are just a number of ways AI is helping to improve patient care and surgical precision by minimising risks.
Top AI-driven innovations in paediatric surgery
AI-assisted diagnostics
AI systems are being trained on a huge amount of medical data such as patient records and medical images, to spot any patterns in the data. For example, it might be noticed that children who have the same combination of symptoms often have the same diagnosis. The AI system can then match a child’s symptoms to the vast amount of similar cases it has been trained on to present a medical professional with a list of likely diagnoses to consider.
A 2023 scientific article looked at research studies from 2015-2023 and summarised that AI has shown great promise in accurately and efficiently diagnosing congenital heart diseases. The studies that were reviewed show that AI-based methods are highly sensitive and specific in their diagnoses.4
The assistance of AI in the diagnosing stage of an illness can help doctors catch rare conditions that they might have overseen and may be able to detect an illness earlier. In other words, AI-assisted diagnostics can help reduce human error.
Surgical robotics
Another way the assistance of AI is improving the medical world is through robotic surgery; the integration of AI in surgical robotics helps the machine make really precise movements as well as enhance the quality of imaging techniques which gives surgeons a more detailed view of the area they’re operating on. Furthermore, during the actual surgery procedure, AI has the ability to analyse what is happening in real time and provide feedback as needed so that surgeons may adjust their surgery process to secure a better outcome.
One real-world example of how AI advancements are enhancing neurosurgery is by accurately identifying and locating brain tumours, using deep learning algorithms that are trained on data from brain MRI scans. The AI system that was trained in this task displayed high performance through the accuracy and efficiency with which it located these brain tumours.5
Through combining the capabilities AI has to support robotic surgery, the future of surgical robotics will be one that includes increased precision and minimise risks, resulting in reduced recovery time.
Personalised treatment plans
AI creates personalised treatment plans by analysing a patient's unique data, like medical history, genetics, and current health conditions. Again, it searches for patterns and insights from similar cases and uses this information to suggest the best treatments. It can also predict how patients might respond to different treatments, helping doctors choose the most effective one. This results in more precise and tailored healthcare, improving outcomes for each individual.
One study from 2021 looked at various possible applications AI could have for paediatric cancer patients by consolidating decision-making through predicting possible outcomes, as well as monitoring cancer progression in children.6 These possibilities are echoed through a number of studies such as this one conducted in 2022 to highlight AI’s potential to automatically and accurately classify different types of leukaemia.7
The benefits of harnessing AI assistance to drive personalised treatment plans would mean fewer false positives when identifying conditions, guiding doctors to implement customised care and therapies, stricter adherence to treatment, and overall enhanced effectiveness of treatments.
Impact, challenges, and future prospects
From the handful and AI innovations discussed in this article that are being applied to paediatric care, we have established that AI assistance is increasing surgical success rates which improves patient recovery time, as well as helps streamline processes, customise treatment plans, and assist medical professionals to consider all possible diagnoses.
Although AI implementation is seen in nearly all sectors of life, it is still a relatively new technological advancement and comes with many challenges and ethical considerations that are yet to be fully ironed out. Specific to AI assistance in healthcare, there are issues surrounding data privacy, potential biases in AI algorithms, as well as the cost and accessibility of it all - especially in paediatric care. There are many questions left unanswered as to how data protection regulations will be adhered to when using patient data to train AI algorithms, as well as defining universal data protection laws in order to exchange patient information globally. And while AI algorithms are trained on huge datasets, these datasets will still not be able to represent all the different subpopulations around the world e.g. think of certain subgroups who might not have access to healthcare and therefore these subgroups have little to no reliable information about possible predisposed health conditions.8 Lastly, one of the biggest hurdles in adopting AI innovations in the paediatric population is the cost factor of making these innovations accessible.9
Despite these challenges, new technologies are constantly emerging and the future of AI in paediatric healthcare is incredibly promising. Looking ahead, we can envision a healthcare landscape where AI tools routinely support paediatric surgeons to ensure faster, safer, and more accurate procedures. But in order for this vision to be achievable, ongoing collaboration between technologists and healthcare professionals needs to take place. By working together, they can ensure these AI systems are progressive while also being practical and safe for everyday use in hospitals - ultimately leading to better healthcare for children.
Conclusion
We have seen how AI in paediatric surgery has begun to reduce recovery times and readmissions by improving diagnostic accuracy and surgical precision through real-time guidance and enhanced imaging. It analyses patient data to predict surgical outcomes and complications, suggesting personalised care plans for a speedier recovery time. Despite challenges like data privacy and cost, ongoing collaboration between technologists and healthcare professionals promises a future where AI routinely supports paediatric healthcare. With so much transformative potential in the innovations being studied so far, the accessibility and reliability of healthcare may very well be redefined for the better in our lifetime. To make this vision a reality, it's crucial to support and invest in AI advancements in healthcare, ensuring better outcomes for all children.
References
- Robotic surgery is safer and improves patient recovery time [Internet]. ScienceDaily. Available from: https://www.sciencedaily.com/releases/2022/05/220515113215.htm
- IBM. What Is Artificial Intelligence (AI)? [Internet]. IBM. 2024. Available from: https://www.ibm.com/topics/artificial-intelligence
- IBM. Artificial Intelligence in Medicine | IBM [Internet]. www.ibm.com. IBM; 2020. Available from: https://www.ibm.com/topics/artificial-intelligence-medicine
- Ejaz H, Tarannum Thyyib, Ibrahim A, Aroob Nishat, Malay J. Role of artificial intelligence in early detection of congenital heart diseases in neonates. Frontiers in digital health [Internet]. 2024 Jan 11;5. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10808664/
- Kalli VDR. Creating an AI-powered platform for neurosurgery alongside a usability examination: Progressing towards minimally invasive robotics. Journal of Artificial Intelligence General science (JAIGS) ISSN:3006-4023 [Internet]. 2024 Apr 19;3(1):363–75. Available from: https://ojs.boulibrary.com/index.php/JAIGS/article/view/125
- Ramesh S, Sukarn Chokkara, Shen T, Major A, Volchenboum SL, Anoop Mayampurath, et al. Applications of Artificial Intelligence in Pediatric Oncology: A Systematic Review. 2021 Dec 1;(5):1208–19.
- Monaghan SA, Li JL, Liu YC, Ko MY, Boyiadzis M, Chang TY, et al. A Machine Learning Approach to the Classification of Acute Leukemias and Distinction From Nonneoplastic Cytopenias Using Flow Cytometry Data. American Journal of Clinical Pathology [Internet]. 2022 Apr 1;157(4):546–53. Available from: https://pubmed.ncbi.nlm.nih.gov/34643210/
- khan B, Fatima H, Qureshi A, Kumar S, Hanan A, Hussain J, et al. Drawbacks of Artificial Intelligence and Their Potential Solutions in the Healthcare Sector. Biomedical Materials & Devices [Internet]. 2023 Feb 8;1(36785697):1–8. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9908503/
- Mei H, Tang S. Robotic-assisted surgery in the paediatric surgeons’ world: Current situation and future perspectives. Frontiers in Pediatrics. 2023 Feb 14;11.