OpenAI’s ChatGPT system sent the topic of artificial intelligence over the roof.
But a lot of professionals across industries, including healthcare, don’t really understand how AI works — especially how different forms of AI work.
Furthermore, there are a lot of acronyms floating around in the tech space: AI (artificial intelligence), ML (machine learning) and now FL (associative learning). But what is the difference between them and how does each relate to healthcare?
For a brief overview of this important topic, Healthcare IT News spoke with Ittai Dayan, CEO and co-founder of Rhino Health. Rhino Health is a platform provider designed to enable developers and researchers to analyze data, create AI models, and deploy them.
Ittai is the author of a diverse clinical association learning study, EXAM (EMR CXR AI Model), published in Natural Medicine last year.
Q. What is AI and how is it used in healthcare today?
ONE. Artificial intelligence refers to the ability of machines to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision making, and language translation. AI systems can learn from experience, adjust to new inputs, and perform human-like tasks without explicit programming.
In healthcare, AI is being used in a number of ways to improve patient outcomes and streamline medical processes. For example, AI-powered diagnostic tools can assist doctors in identifying diseases and conditions based on symptoms, medical history, and other patient data.
AI algorithms can also be used to analyze large amounts of medical data, helping to uncover new insights and treatment options. Additionally, AI can be used to develop personalized treatment plans, monitor patients remotely, and improve the efficiency of clinical trials.
AI is helping healthcare providers make more informed decisions, improve patient outcomes, and deliver more efficient and effective care.
Q. Now, let’s dive in. What is machine learning and what can it be used for in healthcare?
ONE. Machine learning is a subfield of AI that focuses on developing algorithms and statistical models that allow computers to improve their performance on a particular task. In contrast to traditional programming, where rules and logic are clearly defined, machine learning algorithms are designed to automatically improve their performance by learning from data.
There are different types of machine learning, including supervised learning (labels that define ‘basic facts’), unsupervised learning (without labels), and reinforcement learning (machine learning algorithms that learn from ‘experience’. ), each has its own strengths and weaknesses.
In the healthcare sector, machine learning is being used to improve many processes and outcomes. For example, machine learning algorithms can be used to analyze large amounts of medical data, such as electronic health records, to identify patterns and relationships that can inform the diagnosis. develop more effective treatments.
Machine learning can also be used to develop predictive models that can help healthcare providers predict patient outcomes and make more informed decisions. Machine learning is playing an important role in advancing the healthcare sector by enabling more accurate, personalized and effective treatments.
Q. What is affiliate learning and what are its healthcare applications? How is it different from machine learning?
ONE. Association learning is a distributed machine learning technique where each participant has its own data and the model is trained by aggregating updates from these participants without sharing the raw data.
In other words, the data remains on the local device and only the model parameters are transmitted to the central server for aggregation and updating. This approach enables organizations to protect data privacy, security, and ownership while leveraging the benefits of machine learning.
Associative learning and machine learning are related but distinct concepts. Machine learning refers to the development of algorithms and statistical models that allow computers to improve their performance in a particular task through experience.
In contrast, federated learning is a specific type of machine learning that allows multiple participants to collaborate and train a shared model without sharing their raw data.
Associative learning can improve machine learning models in healthcare by enabling use of larger and more diverse datasets while protecting privacy and security. Some of the key ways that associative learning can improve machine learning models in healthcare include:
-
Improve data diversity: Associative learning allows the use of data from multiple sources, including hospitals, clinics, and patients, providing a more diverse dataset to train models. This leads to models that are more generalizable and capable of making more accurate predictions for more patients.
-
Enhanced data privacy and security: By keeping data on local devices, federated learning ensures that sensitive patient data is never exposed or shared between organizations. This helps protect patient privacy and security, and can increase patient trust in technology.
-
More transparency and trust: Federated learning allows data ‘custodians’ to maintain control over their data and provides a simple way for them to enforce contracts and ensure transparency across the board. ‘lifecycle’ set of data.
Q. Please tell us about your EXAM affiliated learning research and what can healthcare IT leaders in healthcare learn from this research?
ONE. The EXAM study is a research project – led by me and Dr. Mona Flores, Nvidia’s global head of medical AI – that was published in the September 2021 issue of Nature Medicine. demonstrate the feasibility and benefits of associate learning in the healthcare sector.
A model was developed using local as well as network data to predict the outcomes of patients presenting to the emergency department with respiratory complaints.
EXAM research has demonstrated that federated learning can enable hospitals to collaborate and provide federated access to data without compromising patient privacy and security.
The study shows that associative learning can improve the performance of the predictive model, creating a globally linked model that is better than any local model, and that proves the level of generalizability. high fidelity to data not seen in subsequent validation studies.
This therefore demonstrates that associative learning has the potential to change the way hospitals collaborate to improve patient outcomes.
EXAM’s results show that there is a way to overcome some of the major challenges associated with data sharing in healthcare, such as privacy, security, and data ownership. The study provides a road map for how healthcare organizations can use federated learning to improve patient outcomes while protecting privacy and security.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email the writer: [email protected]
Healthcare IT News is a publication of HIMSS Media.
