Artificial intelligence and machine learning already are having an impact in a sector that concerns all of us, healthcare. AI/ML can be applied to various types of healthcare data (structured and unstructured). It aims to mimic human cognitive functions and is bringing a paradigm shift to healthcare, powered by increasing availability of healthcare data and rapid progress of analytics techniques. The early diagnosis of diseases can improve the prognosis and chance of survival significantly, as it can promote timely clinical treatment to patients.
Ms. Dyuti Lal spoke and presented some of the important highlights of the implementation on how AI/ML can typically be categorized as aiding with one or more of the following: Keeping Well, Early Detection, Diagnosis, Decision-Making, Treatment, Research, etc along with few interesting case studies at Machine Conference 2019, Singapore.
What is AI ???
Artificial intelligence (AI) is an area of computer science that emphasizes the creation of intelligent machines that work and react like humans.
What is Machine Learning ???
Machine Learning is the field of study that gives computers the capability to learn without being explicitly programmed, relying on patterns and inference instead. Machine learning algorithms build a mathematical model based on sample data, known as “training data”, in order to make predictions or decisions without being explicitly programmed to perform the task.
What is Deep learning ???
Deep learning is a subset of machine learning where artificial neural networks, algorithms inspired by the human brain, learn from large amounts of data. Similarly to how we learn from experience, the deep learning algorithm would perform a task repeatedly, each time tweaking it a little to improve the outcome. We refer to ‘deep learning’ because the neural networks have various (deep) layers that enable learning.
Deep learning allows machines to solve complex problems even when using a data set that is very diverse, unstructured and inter-connected. The more deep learning algorithms learn, the better they perform.
Some of the activities computers with artificial intelligence are designed for include:
- Speech recognition
- Problem solving
Some of the Applications of Machine Learning would be in:
1.Fraud Detection (Credit Card Transactions/Insurance Claim Genuine/Fraudulent in BFSI)
2.Recommendation System (Product/movie recommendations on Amazon/Netflix etc)
3.Detection of Disease at early stages (eg. Cancer is benign or malign in Healthcare)
4. Image Recognition & Categorization (eg.Airbnb uses machine learning to help categorize its listing photos)
5. Self Driving Cars etc.
Applications of Deep Learning:
- Virtual Assistants such as Alexa, Siri, Cortana
- Translations between languages
- Healthcare for disease and tumor diagnosis
- Facial Recognition
- Personalized Shopping & Entertainment
AI in Healthcare:
Many industries have been disrupted by the influx of new technologies in the Information Age. Healthcare is no different. Particularly in the case of automation, machine learning, and artificial intelligence (AI), doctors, hospitals, insurance companies, and industries with ties to healthcare have all been impacted – in many cases in more positive, substantial ways than other industries.
According to a 2016 report from CB Insights, about 86% of healthcare provider organizations, life science companies, and technology vendors to healthcare are using artificial intelligence technology. By 2020, these organizations will spend an average of $54 million on artificial intelligence projects.
So what solutions are they most commonly implementing? Here are 10 common ways AI is changing healthcare now and will in the future.
1. Managing Medical Records and Other Data
Since the first step in health care is compiling and analyzing information (like medical records and other past history), data management is the most widely used application of artificial intelligence and digital automation. Robots collect, store, re-format, and trace data to provide faster, more consistent access.
2. Doing Repetitive Jobs
Analyzing tests, X-Rays, CT scans, data entry, and other mundane tasks can all be done faster and more accurately by robots. Cardiology and radiology are two disciplines where the amount of data to analyze can be overwhelming and time consuming. Cardiologists and radiologists in the future should only look at the most complicated cases where human supervision is useful.
3. Treatment Design
Artificial intelligence systems have been created to analyze data – notes and reports from a patient’s file, external research, and clinical expertise – to help select the correct, individually customized treatment path.
4. Digital Consultation
Apps like Babylon in the UK use AI to give medical consultation based on personal medical history and common medical knowledge. Users report their symptoms into the app, which uses speech recognition to compare against a database of illnesses. Babylon then offers a recommended action, taking into account the user’s medical history.
5. Virtual Nurses
The startup Sense.ly has developed Molly, a digital nurse to help people monitor patient’s condition and follow up with treatments, between doctor visits. The program uses machine learning to support patients, specializing in chronic illnesses.
In 2016, Boston Children’s Hospital developed an app for Amazon Alexa that gives basic health information and advice for parents of ill children. The app answers asked questions about medications and whether symptoms require a doctor visit.
6. Medication Management
The National Institutes of Health have created the AiCure app to monitor the use of medication by a patient. A smartphone’s webcam is partnered with AI to autonomously confirm that patients are taking their prescriptions and helps them manage their condition. Most common users could be people with serious medical conditions, patients who tend to go against doctor advice, and participants in clinical trials.
7. Drug Creation
Developing pharmaceuticals through clinical trials can take more than a decade and cost billions of dollars. Making this process faster and cheaper could change the world. Amidst the recent Ebola virus scare, a program powered by AI was used to scan existing medicines that could be redesigned to fight the disease.
The program found two medications that may reduce Ebola infectivity in one day, when analysis of this type generally takes months or years – a difference that could mean saving thousands of lives.
8. Precision Medicine
Genetics and genomics look for mutations and links to disease from the information in DNA. With the help of AI, body scans can spot cancer and vascular diseases early and predict the health issues people might face based on their genetics.
9. Health Monitoring
Wearable health trackers – like those from FitBit, Apple, Garmin and others – monitors heart rate and activity levels. They can send alerts to the user to get more exercise and can share this information to doctors (and AI systems) for additional data points on the needs and habits of patients.
10. Healthcare System Analysis
In the Netherlands, 97% of healthcare invoices are digital. A Dutch company uses AI to sift through the data to highlight mistakes in treatments, workflow inefficiencies, and helps area healthcare systems avoid unnecessary patient hospitalizations.
These are just a sample of the solutions AI is offering the healthcare industry. As innovation pushes the capabilities of automation and digital workforces, from providers like Novatio, more solutions to save time, lower costs, and increase accuracy will be possible.
Branches of AI in Medicine
AI in medicine has 2 major branches:
The virtual branch includes informatics approaches from deep learning information management to control of health management systems, including electronic health records, and active guidance of physicians in their treatment decisions
Supervised Machine Learning:
Supervised learning is a type of machine learning algorithm that uses a known dataset (called the training dataset) to make predictions. The training dataset includes input data and response values. From it, the supervised learning algorithm seeks to build a model that can make predictions of the response values for a new dataset.
Supervised learning includes two categories of algorithms:
- Classification: for categorical response values, where the data can be separated into specific “classes”
Common classification algorithms include:
- Support vector machines (SVM)
- Neural networks
- Naïve Bayes classifier
- Decision trees
- Discriminant analysis
- Nearest neighbors (kNN)
- Regression: for continuous-response values
Common regression algorithms include:
- Linear regression
- Nonlinear regression
- Generalized linear models
- Decision trees
- Neural networks
Unsupervised Machine Learning:
Unsupervised learning is a type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses.
The most common unsupervised learning method is cluster analysis, which is used for exploratory data analysis to find hidden patterns or grouping in data. The clusters are modeled using a measure of similarity which is defined upon metrics such as Euclidean or probabilistic distance.
Reinforcement Machine Learning
Reinforcement Learning is a type of Machine Learning, and thereby also a branch of Artificial Intelligence. It allows machines and software agents to automatically determine the ideal behavior within a specific context, in order to maximize its performance.
The physical branch is best represented by robots used to assist the elderly patient or the attending surgeon. Also embodied in this branch are targeted nano robots, a unique new drug delivery system
- Technological limitations:
At best, current technology – meaning various machine learning methods – is able to reach artificial narrow intelligence (ANI) in various fields. Yet, that is developing at an incredible speed. These narrowly intelligent programs defeat humans in specific tasks, such as IBM’s supercomputer Deep Blue winning at chess but unlike human world champions, these algorithms are not capable of also driving cars or creating art. Solving those other tasks requires other narrow programs to be built, and it is an immense challenge. Yet, there is incredible growth in computers’ ability to understand images and video – a field called computer vision – as well as text in the frames of natural language processing. The former is extensively utilized now in healthcare, for example in the field of medical imaging.
2. Medical limitations
To avoid over-hyping the technology, the medical limitations of present-day ANI also have to be acknowledged. In the case of image recognition and using machine learning and deep learning algorithms for the purposes of radiology, there is the risk of feeding the computer not only with thousands of images but also underlying bias.
On the other hand, streamlining and standardizing medical records in such a way that algorithms can make sense of them mean another huge limitation in introducing ANI to hospital departments for doing administrative tasks. There are many hospitals where doctors still scribble their notes on patients’ files. How should the computer make sense of such notes if even the person who wrote that cannot read it two weeks later?
3. Ethical challenges
Yet, medical as well as technological limitations of A.I. as well as ANI will still be easier to overcome than ethical and legal issues. Who is to blame if a smart algorithm makes a mistake and does not spot a cancerous nodule on a lung X-ray? To whom could someone turn when A.I. comes up with a false prediction? Who will build in safety features so A.I. will not turn on humans? What will be the rules and regulations to decide on safety?
These complex ethical and legal questions should be answered if we want to reach the stage of AGI safely and securely. Moreover, ANI and at a certain point, AGI, should be implemented cautiously and gradually in order to give time and space for mapping the potential risks and downsides. Independent bioethical research groups, as well as medical watchdogs, should monitor the process closely. This is exactly what the Open AI Foundation does on a broader scale. It is a non-profit A.I. research company, discovering and enacting the path to safe artificial general intelligence. Their work is invaluable, as they are doing long-term research, and may help in setting up ethical standards on how to use A.I. on micro and macro levels. Perhaps also in the healthcare sector.
4. Misconceptions and overhyping
Over hyping the capabilities of A.I. through marketing tactics and oversimplified media representations does not help but destroy a healthy image about how A.I. could contribute to healthcare. It also adds to the fog of confusion and misconceptions which need to be cleared up when we want to implement the technology successfully into our healthcare systems.
5. Human rejection
The fears around A.I. are understandable as so few of us actually understand how the technology works down to the detail. And what we don’t understand, we tend to reject. Even more so, if thought-leaders or the media also tend to treat the issue with exaggeration and extremities. And although it will take time to get accustomed to the technology, we recommend everyone to be open-minded and familiarize with the concept of using A.I. in everyday life.
Case Studies in Part 2.