Connect with us

Health Care

How AI Bias Happens – and How to Eliminate It

Published

on

Artificial intelligence holds great promise for healthcare, and it is already being put to use by many forward-looking hospitals and health systems.

One challenge for healthcare CIOs and clinical users of AI-powered health technologies is the biases that may pop up in algorithms. These biases, such as algorithms that improperly skew results because of race, can compromise the ultimate work of AI – and clinicians.

We spoke recently with Dr. Sanjiv M. Narayan, co-director of the Stanford Arrhythmia Center, director of its Atrial Fibrillation Program and professor of medicine at Stanford University School of Medicine. He offered his perspective on how biases arise in AI – and what healthcare organizations can do to prevent them.

Q. How do biases make their way into artificial intelligence?

A. There is an increasing focus on bias in artificial intelligence, and while there is no cause for panic yet, some concern is reasonable. AI is embedded in systems from wall to wall these days, and if these systems are biased, then so are their results. This may benefit us, harm us or benefit someone else.

A major issue is that bias is rarely obvious. Think about your results from a search engine “tuned to your preferences.” We already are conditioned to expect that this will differ from somebody else’s search on the same topic using the same search engine. But, are these searches really tuned to our preferences, or to someone else’s preferences, such as a vendor? The same applies across all systems.

Bias in AI occurs when results cannot be generalized widely. We often think of bias resulting from preferences or exclusions in training data, but bias can also be introduced by how data is obtained, how algorithms are designed, and how AI outputs are interpreted.

How does bias get into AI? Everybody thinks of bias in training data – the data used to develop an algorithm before it is tested on the wide world. But this is only the tip of the iceberg.

All data is biased. This is not paranoia. This is fact. Bias may not be deliberate. It may be unavoidable because of the way that measurements are made – but it means that we must estimate the error (confidence intervals) around each data point to interpret the results.

Think of heights in the U.S. If you collected them and put them all onto a chart, you’d find overlapping groups (or clusters) of taller and shorter people, broadly indicating adults and children, and those in between. However, who was surveyed to get the heights? Was this done during the weekdays or on weekends, when different groups of people are working?

If heights were measured at medical offices, people without health insurance may be left out. If done in the suburbs, you’ll get a different group of people compared to those in the countryside or those in cities. How large was the sample?

Bias in training data is the bias that everybody thinks about. AI is trained to learn patterns in data. If a particular dataset has bias, then AI – being a good learner – will learn that too.

A now classic example is Amazon. Some years ago, Amazon introduced a new AI-based algorithm to screen and recruit new employees. The company was disappointed when this new process did nothing to help diversity, equity and inclusion.

“All data is biased. This is not paranoia. This is fact.”

Dr. Sanjiv M. Narayan, Stanford University School of Medicine

When they looked closely, it turned out that that the data used for training came from applications submitted to Amazon primarily from white men over a 10-year period. Using this system, new applicant resumes were downgraded if they contained the terms “women’s” or “women’s colleges.” Amazon stopped using this system.

On another front, AI algorithms are designed to learn patterns in data and match them to an output. There are many AI algorithms, and each has strengths and weaknesses. Deep learning is acknowledged as one of the most powerful today, yet it performs best on large data sets that are well labeled for the precise output desired.

Such labeling is not always available, and so other algorithms are often used to do this labeling automatically. Sometimes, labeling is done not by hand, but by using an algorithm trained for a different, but similar, task. This approach, termed transfer learning, is very powerful. However, it can introduce bias that is not always appreciated.

Other algorithms involve steps called auto-encoders, which process large data into reduced sets of features that are easier to learn. This process of feature extraction, for which many techniques exist, can introduce bias by discarding information that could make the AI smarter during wider use – but that are lost even if the original data was not biased.

There are many other examples where choosing one algorithm over another can modify results from the AI.

Then there is bias in reporting results. Despite its name, AI is typically not “intelligent” in the human sense. AI is a fast, efficient way of classifying data – your smartphone recognizing your face, a medical device recognizing an abnormal pattern on a wearable device or a self-driving car recognizing a dog about to run in front of you.

The internal workings of AI involve mathematical pattern recognition, and at some point all of this math has to be put into a bin of Yes or No. (It’s your face or not, it’s an abnormal or normal heart rhythm, and so on.) This process often requires some fine-tuning. This may be to reduce bias in data collection, in the training set, in the algorithm, or to attempt to broaden the usefulness.

For instance, you may decide to make your self-driving car very cautious, so that if it senses any disturbance at the side of the road it alarms “caution,” even if the internal AI would have not sounded the alarm.

Q. What kind of work are you currently doing with AI?

A. I am a professor and physician at Stanford University. I treat patients with heart conditions, and my lab has for a long time done research into improving therapy in individual patients using AI and computer methods to better understand disease processes and health.

In cardiology, we are fortunate in having many ways to measure the heart that increasingly are available as wearable devices and that can directly guide treatment. This is very exciting, but also introduces challenges. One major issue that is emerging in medicine is AI bias.

Bias in medical AI is a major problem, because making a wrong diagnosis or suggesting [the] wrong therapy could be catastrophic. Each of the types of bias I have described can apply to medicine. Bias in data collection is a critical problem. Typically, we only have access to data from patients we see.

However, what about patients without insurance, or those who only choose to seek medical attention when very sick? How will AI work when they ultimately do present to the emergency room? The AI may have been trained on people who were less sick, younger or of different demographics.

Another interesting example involves wearables, which can tell your pulse by measuring light reflectance from your skin [photo-plethysmography]. Some of these algorithms are less accurate in people of color. Companies are working on solutions that address this bias by working on all skin tones.

Other challenges in medical AI include ensuring accuracy of AI systems (validation), ensuring that multiple systems can be compared for accuracy, which ideally would use the same testing data. But this may be proprietary for each specific system – and ensuring that patients have access to their data. The Heart Rhythm Society recently called for this “transparent sharing” of data.

Q. What is one practice for keeping biases out of AI?

A. Understanding the various causes of bias is the first step in the adoption of what is sometimes called effective “algorithmic hygiene.” An essential practice is to ensure as much as possible that training data are representative.

Representative of what? No data set can represent the entire universe of options. Thus, it is important to identify the target application and audience upfront, and then tailor the training data to that target.

A related approach is to train multiple versions of the algorithm, each of which is trained to input a dataset and classify it, then repeat this for all datasets that are available. If the output from classification is the same between models, then the AI models can be combined.

A similar approach is to input the multiple datasets to the AI, and train it to learn all at once. The advantage of this approach is that the AI will learn to reinforce the similarities between input datasets, and yet generalize to each dataset.

As AI systems continue to be used, one tailored design is to update their training dataset so that they are increasingly tailored to their user base. This can introduce unintended consequences. First, as the AI becomes more and more tailored to the user base, this may introduce bias compared to the carefully curated data often used originally for training.

Second, the system may become less accurate over time because the oversight used to ensure AI accuracy may no longer be in place in the real world. A good example of this is the Microsoft ChatBot, which was designed to be a friendly companion but, on release, rapidly learned undesirable language and behaviors, and had to be shut down.

Finally, the AI is no longer the same as the original version, which is an issue for regulation of medical devices as outlined in the Food and Drug Administration guidelines on Software as a Medical Device.

Q. What is another best practice for preventing AI bias?

A. There are multiple approaches to eliminate bias in AI, and none are foolproof. These range from approaches to formulate an application so that it is relatively free of bias, to collecting data in a relatively unbiased way, to designing mathematical algorithms to minimize bias.

The technology of AI is moving inexorably toward greater integration across all aspects of life. As this happens, bias is more likely to occur through the compounding of complex systems but also, paradoxically, less easy to identify and prevent.

It remains to be seen how this field of ethical AI develops and whether quite different approaches are developed for highly regulated fields such as medicine, where transparency and explainable AI are of critical importance, and other endeavors.

Twitter: @SiwickiHealthIT
Email the writer: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.

Source Here: healthcareitnews.com

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Health Care

Singapore’s Public Health System Rolling Out the Clinician’s ZEDOC Platform

Published

on

Singapore’s health tech agency Integrated Health Information Systems has partnered with Auckland-headquartered digital health firm The Clinician to deploy a patient-reported outcome and experience measures platform across the island state’s public healthcare system.

WHAT IT’S FOR

The Clinician’s ZEDOC platform, the company describes, assists healthcare providers in managing patient-generated health data outside the hospital through digitisation. Integrated with HIS, the system supports timely exchange of health data and information between providers and patients, including subjective PROMs and PREMs, objective wearable device data, and other communication or educational materials. By streamlining the digital collection of critical health data, ZEDOC is able to render real-time, actionable information crucial for improving health outcomes and experiences.

The partners are working on multiple ZEDOC integrations with existing health information systems (HIS). A privacy-preserving hybrid infrastructure has been implemented which ensures that all personally identifiable information stays within the IHiS’s private health cloud while all anonymised health data are collected through a secure commercial cloud platform.

WHY IT MATTERS

Singapore intends to measure and improve health outcomes and patient experience with the rollout of The Clinician’s ZEDOC platform. Their partnership will “bolster patient engagement and enable clinicians to more effectively assess patients’ health status before, during and after receiving a health service – closing the loop when they are outside the hospital,” said The Clinician CEO Dr Ron Tenenbaum. It will also allow providers to deliver “more holistic and personalised care for patients by taking into account their perspectives for the first time,” he added.

To demonstrate the benefit of routine collection and analysis of PROMs, The Clinician shared that this has resulted in over 50% reduction in 90-day complications for hip and knee surgery patients in one study and a five-month improvement in the survival of cancer patients in another.

Among benefits for care providers, the ZEDOC integration will replace existing paper-based forms with an integrated digital platform that automates data capture, as well as benchmark outcomes across providers to reduce variability and waste. For patients, they can become more involved in the treatment decision-making and be informed early of health risks and warning signs.

THE LARGER TREND

Last month, Cabrini Health and The Alfred, two of the largest healthcare providers in the Australian state of Victoria, deployed the ZEDOC platform to automate the collection and analysis of health data from colorectal cancer patients. The installation is said to adhere to the colorectal cancer standards outlined by the International Consortium of Health Outcomes Measurement.

Original Post: healthcareitnews.com

Continue Reading

Health Care

EU Analysis Highlights Digital Health Lessons From COVID-19

Published

on

An EU analysis has outlined the effect of COVID-19 on healthcare systems in Europe and the role of digital innovation in building their resilience.

Experts from the Organisation for Economic Co-operation and Development (OECD) and the European Observatory have published a set of 29 country health profiles, covering all EU member states, as well as Iceland and Norway. A companion report also highlights a selection of cross-country trends.

Speaking at a virtual launch event on Monday (13 December), Josep Figueras, director, European Observatory, highlighted two main lessons learnt from the use of technology in the pandemic.

Using telemedicine as an example of digital health innovation, he said the number of teleconsultations had increased in all EU countries during 2020. However in some countries, such as France, teleconsultations had decreased when lockdowns ended.

“The key issue here is how we harness and sustain innovation – how we make sure that these improvements in the use of telemedicine (as an illustration of the use of other digital technologies) can be maintained and sustained to increase the effectiveness of the health system,” Figueras said.

He also highlighted that the technology for telemedicine and other innovations was already available in many European countries before the pandemic but was not being used.

Figueras asked: “What did we do within the pandemic that literally within a couple of weeks, we got all this telemedicine in place?”

To sustain the use of telemedicine and other health technologies, he said it was important to look at the regulatory measures, financial incentives, training and changes in culture needed.

“Something the pandemic has taught us loudly and clearly is the importance of digital innovation – not only the new technologies, but the ability to implement them,” Figueras added.

WHY IT MATTERS

The State of Health in the EU cycle is a two-year process initiated by the European Commission in 2016, designed to improve country-specific and EU-wide knowledge in healthcare.

It aims to gather data and in-depth analyses on health systems and make the information accessible to policy makers and stakeholders.

THE LARGER CONTEXT

During the pandemic, digital tools have been used in the EU to boost public health measures such as the implementation of the EU Digital COVID Certificate, vaccination booking systems, and cross-border interoperability for contact-tracing apps.

There has also been investment in EU-wide COVID recovery initiatives such as the EU4Health programme.

ON THE RECORD

Maya Matthews, head of unit performance, European Commission said: “COVID-19 illuminated the fact that in many European countries we do not have a strong public health system. We cannot do testing and tracing. Even surveillance is done sometimes in a very fragmented fashion.

“I think if one thing comes out of COVID-19, it’s to say that public health matters – that public health is a very important part of health systems and has not really received the attention it deserves.”

Source Here: healthcareitnews.com

Continue Reading

Health Care

Clinical Messaging Platform Hospify to Close, Bupa Arabia Invests in Global Ventures, and More News Briefs

Published

on

Clinical messaging platform Hospify to close

British healthtech startup Hospify has announced it will close its secure clinical messaging platform on 31 January 2021.

Hospify said it suffered a decline in demand after the government suspended the UK 2018 Data Protection Act in relation to healthcare last year for the duration of the COVID-19 pandemic.

It also cited difficulties caused by “post-Brexit uncertainties surrounding the future of the UK’s data adequacy agreement with the EU”.

A statement from the Hospify team says: “It’s a sad end to a wonderful vision, a vision of universal health care communication that was both free of data exploitation and free at the point of use.”

Insurance giant Bupa Arabia invests in Global Ventures

UAE-based international venture capital firm Global Ventures has announced new investment from Bupa Arabia, the leading health insurance company in the region.

Bupa Arabia’s participation in Global Ventures Fund II as strategic partner aims to foster the healthcare ecosystem in the region and particularly in Saudi Arabia.

The investment is part of the Bupa Arabia’s strategy to participate and invest in disruptive healthcare and insurance technologies, amongst other targeted growth sectors.

Noor Sweid, Global Ventures founder and general partner, said: “Bupa Arabia shares our outlook and ambition on the digital health sector, and its potential for technology and innovation to deliver long-term economic benefits particularly in emerging markets.”

Liverpool Heart and Chest Hospital achieves EMRAM Stage 6

Specialist NHS trust Liverpool Heart and Chest Hospital (LHCH) has been awarded Stage 6 of the EMRAM, or Electronic Medical Record Adoption Model, by HIMSS.

The EMRAM measures the adoption and maturity of a health facility’s inpatient EMR capabilities from 0 to 7. Achieving Stage 6 means the trust has established clear goals for improving safety, minimising errors, and recognising the importance of healthcare IT.

Kate Warriner, chief digital and information officer said: “Digital excellence must be the cornerstone if we are to continually improve the care that we provide for our patients in the years ahead. Therefore, whilst we are rightly proud of this achievement, we have ambitions for further pioneering innovation and advancing our use of technology to become a Stage 7 hospital.”

More than $110m raised by Sheba’s ARC Innovation Center

Israel’s Sheba Medical Center has announced that six companies from its Accelerate Redesign Collaborate (ARC) Innovation Center raised more than $110 million (EUR97.2m) in 2021.

ARC brings new technologies into the hospital and community ecosystem focusing on digital health technologies including precision medicine, big data, artificial intelligence (AI), predictive analytics, telemedicine and mobile health.

Sheba MedTech startups receiving investments this year included: Aidoc, BELKIN Laser, Starget Pharma Append Medical, Innovalve Bio Medical and TechsoMed.

Professor Eyal Zimlichman, ARC director and founder, said: “The ARC Innovation Center has been focusing on ground-breaking, innovative technologies with a prime directive to redesign healthcare.”

Konica Minolta named as part of NHS Digital Documents Solutions framework

Konica Minolta Business Solutions (UK) Ltd has been named as one of 46 suppliers on the new ?5 billion Digital Documents Solutions framework.

The firm will provide solutions across five key areas: internal print, external print, digital mail room, scanning and electronic document management solutions.

Jason Barnes, head of public sector, Konica Minolta, said: “Having been chosen through a competitive tender process, we are especially pleased to be newly appointed to the LPP framework, which deepens and furthers our reach into the NHS health sector.”

Original Source: healthcareitnews.com

Continue Reading

Trending

RLER.com