Updated: 3 days ago
In this article, Artificial Intelligence in Surgery: Promises and Perils, four core AI subfields were introduced.
Machine Learning (ML)
Natural Language Processing (NLP)
Artificial neural network (ANN)
Computer Vision (CV)
Artificial intelligence (AI) is often referred to as computer simulation of human intelligence.
AI is an interdisciplinary field that encompasses computer science, mathematics, biology, psychology, linguistics, logic, and philosophy.
The term “artificial intelligence” was coined in 1955 by John McCarthy, a professor at Dartmouth.
1. Machine learning
Machine learning (ML) allows machines to learn and make predictions by recognizing models.
ML is how machines learn.
ML outperformed logistic regression for predicting surgical site infection.
2. Natural language processing
Natural Language Processing (NLP) is a subdomain that focuses on the computer’s ability to understand human language.
NLP is the way machines understand human language.
The NLP detects adverse events and post-operative complications by identifying words and phrases in operational reports and progress notes.
3. Artificial neural network
Artificial neural networks (ANN), a subdomain of ML, are inspired by biological nervous systems.
Deep learning is the process of using multi-layer artificial neural networks to learn complex models in data.
ANN is a simulation of the human nervous system.
ANN has surpassed more traditional approaches to risk prediction (predicting the severity of pancreatitis, predicting hospital mortality after an open abdominal aortic aneurysm repair…).
4. Computer vision
Computer vision (CV) describes machine understanding of images, videos, objects, and scenes.
CV is how machines understand images and videos.
The important application of CV in surgery includes computer-assisted diagnosis, image-guided surgery, virtual colonoscopy and predictive video analysis.
AI and Machine Learning (ML) have become indispensable in surgical practices.
While these technologies offer considerable potential, surgeons must remain vigilant about the practical and ethical AI dilemmas.
AI and ML can serve as tools that augment the very skills surgeons themselves already possess.
For example, an algorithm trained to identify pneumonia can assist a clinician by analyzing medical images.
Such AI tools ultimately perform tasks that require validation and confirmation from the physician.
The responsibility of diagnosis and decision-making still rests with the physician.
As AI capabilities progress, especially in predictive analytics and data analysis, physicians may encounter a dilemma.
Surgeons might find themselves relying on clinical information provided by AI/ ML without the means to verify its veracity independently.
This raises a fundamental concern:
What happens if the algorithm makes an error?
Who bears the liability for such mistakes?
Is it the physician? The hospital that bought the software? Or the company that produced the software?
Should AI algorithms be subjected to the same approval processes as traditional treatments by the U.S. Food and Drug Administration (FDA)?
Finding the balance between innovation and patient safety is important.
Another crucial ethical challenge is patient consent.
Is it necessary for patients to give explicit consent for AI involvement in their healthcare and use their data in training machine learning algorithms?
Only through responsible innovation and implementation can we harness the full potential of AI and ML in surgery while respecting the principles of medical ethics and patient safety.
One emerging application of artificial intelligence (AI) is the evaluation of surgical technical skills.
Many surgeons are evaluated on complication rates, mortality rates, length of stay, blood loss, patient’s length of recovery, and recurrence rates. However, objectively evaluating the technical competence of a surgeon can be challenging.
Both artificial intelligence (AI), machine learning (ML), and deep learning are being used to assess technical skills in surgery, using computer vision (VC).
Computer vision is a subset of artificial intelligence. It is how computers can “see”.
Computer vision enables computers to derive informations from digital images and videos.
Computer vision can interpret and analyze information of data collected from surgeries performed with laparoscopic or robotic surgery.
A recent study used AI to identify operative steps in laparoscopic sleeve gastrectomy and found that quantitative data can be obtained from surgical videos with 85.6% accuracy using artificial intelligence.
AI can assess surgical skills in a variety of ways:
The use of electromagnetic sensors attached to instruments
Hand-mounted eye trackers
Force and sensors attached to surgical instruments
Direct capture from the robot
Robotic instrument vibrations can be measured to determine how forcefully the instruments are handled.
AI can objectively collect data such as:
The number of times an instrument comes into contact with certain structures.
Eye trackers can determine the surgeon’s object of focus.
This data can be used to assess a surgeon’s skill or experience.
Collecting this data and using machine learning to analyze it provides insight into a surgeon’s strengths and weaknesses. It can help identify which skills and maneuvers are important for good patient outcomes and efficient procedure length.
Surgical technical skill can be evaluated by technologies that are already built into surgical equipment, such as the da Vinci Systems recording device.
One outcome that is often used to assess surgical skill is the length of the total procedure being performed.
Using recordings of laparoscopic or robotic surgeries, machine learning can be implemented to analyze the time it takes to perform critical tasks during the surgery, not simply the overall procedure time.
Pauses during the surgery that are considered flow disturbances can be evaluated.
Each step of the surgical procedure can be analyzed, and the time it takes to complete various phases of the operation can be compared.
Using these algorithms, experienced surgeons can be differentiated from beginners within the first 10 seconds of starting a task with 90% accuracy.
Surgical robotic systems provide valuable data that can be utilized by AI to objectively evaluate a surgeon’s technical skill.
These algorithms can also detect patterns that lead to better outcomes, which may help in training future surgeons.