Are Machines The Doctors Of The Future?
AI in Healthcare
This tagline seems to be a bit abhorrent but it will be the future of tomorrow.
Researchers all over the world are working on ML models to predict the medical fatalities way before they even occur. Some of the models are even adept as compared to doctors in identifying the ailments.
In the field of healthcare, a team of researchers of Stanford University led by Andrew Ng (yes yes, the author of the most famous course on ML) proved that the ML model could identify heart arrhythmias from ECG with better efficiency than an expert.
AI has proven to be better at key healthcare tasks like disease diagnosis, where the predefined algorithms outperformed the radiologists at spotting anomalies (like malignant tumors). [1]
Here is a brief overview of how AI-based technology is revolutionizing the healthcare industry.
Machine Learning in Healthcare
We live in the era of data. So, why not put this data to use and predict our future.
ML is used in the medical field for precision medicine — predicting treatment strategies that are likely to benefit a patient taking into consideration their medical history. Supervised learning is used for this prediction as the ML model is trained by the dataset (medical history of the patient) and then predicts the outcome.
Neural Networks in healthcare
A technology that has been long discovered in the 1960s and being researched on, to date, is being used for categorization problems. In the initial years of its discovery, the neural network was assumed to be inefficient since it required a lot of data for computation. But now that we live in a connected world and have access to datasets, neural network emerges out as a potential solution in healthcare.
This technology views the problem statement in categories of inputs, outputs, and weights/features that link inputs with outputs. It works in a similar way as neurons process signals.
Deep Learning in Healthcare
Simply put, the neural network model with many layers of feature/variables makes this technology. It’s widely used in the recognition of malignant lesions in radiology images.
The methodology is extremely beneficial in the field of radiomics- extracting features from radiographic medical images using data-characterization algorithms. With the use of deep learning and radiomics, health experts have been able to uncover disease characteristics that failed to be perceived by the naked eye.
Physical Robots
The “robot” is a common term today. Most of the applications use robots in their core works and are especially beneficial in situations where human intervention can bring risk. In the field of healthcare, robots are now being introduced with AI capabilities being embedded inside their operating system.
While the USA deploys surgical robots from the 2000s to date, this tech has proven to bring extra capabilities to surgeons by expanding their horizon of view to precisely operate upon entangled tissues.
Crucial surgeries of gynecologic, prostate, and head and neck surgery are being done with the assistance of surgical robots.
But wait!
Are the patients ready to believe a machine over a human? — this seems to be a big question.
It’s interesting to know that while machines are becoming skillful in taking decisions and give desirable results, they still lack human emotions and judgemental skills.
Mathematically,
Machines ≠ Humans
“In a survey of more than 300 clinical leaders and healthcare executives, more than 70% of the respondents reported having less than 50% of their patients highly engaged and 42% of respondents said less than 25% of their patients were highly engaged” [3]
Thus, patient involvement and cooperation could be the last milestone in achieving the benefits of an AI system in healthcare.
Conclusion
One has to agree that the recent advancement made by technology in healthcare has brought about a revolutionary change in traditional medical practices. AI systems will never be able to replace human experts (at least not anytime soon) but will indeed supplement their efforts for providing care to patients. Many researchers believe that in the next decade, healthcare will use more AI than we expect it today for the betterment of society.
References and further reading:
- [1] Davenport, Thomas, and Ravi Kalakota. “The potential for artificial intelligence in healthcare.” Future healthcare journal 6.2 (2019): 94.
- [2] https://www.technologyreview.com/search/?s=artificial%20Intelligence
- [3] Davenport TH, Hongsermeier T, Mc Cord KA. Using AI to improve electronic health records. Harvard Business Review 2018. https://hbr.org/2018/12/using-ai-to-improve-electronic-health-records
