​Melanie Cole, MS (Host): Breast cancer is one of the most common cancers affecting women, and early detection can make a life-saving difference. Today, we're talking about a new tool that's making headlines in healthcare, artificial intelligence or AI. What role is it really playing in finding breast cancers earlier, making treatment more personalized and supporting doctors in their decisions. We're here today to learn how AI is shaping the future of breast cancer care.

Welcome to Back to Health, your source for the latest in health, wellness, and medical care, keeping you informed so you can make informed healthcare choices for yourself and your whole family. Back to Health features conversations about trending health topics and medical breakthroughs from our team of world-renowned physicians at Weill Cornell Medicine. I'm Melanie Cole. And I'm joined today by Dr. Magdalena Lombardi Plasilova. She's an associate attending surgeon at New York Presbyterian Hospital, Weill Cornell Medical Center, and an Associate Professor of Clinical Surgery at Weill Cornell Medical College - Cornell University.

Dr. Plasilova, thank you so much for joining us today. To start out, what do we mean when we say artificial intelligence? And then, what do we mean when we say artificial intelligence in breast cancer diagnosis and treatment? Connect the dots here for us.

Dr. Magdalena Lombardi Plasilova: Hi. Thank you for having me today. So, AI, artificial intelligence, in breast cancer diagnosis and treatment essentially refers to the use of advanced computational algorithms, particularly machine learning and deep learning to analyze very large medical data in the breast field. It includes imaging, pathology and genomic and patient information. And all this helps to assist clinicians in detecting, diagnosing, and planning breast cancer treatment. AI helps enhance accuracy, efficiency, personalization, basically make things much better.

Melanie Cole, MS: This is such an interesting topic, Doctor. I mean, really, it seems to be the future now. How is it currently being used? As we start, let's talk about how it's currently being used, and we'll talk about what we see in the future as well. But in breast cancer screening specifically, does it work alongside mammograms or other imaging tests like ABUS or ultrasound? How is it working in screening? How are we using it?

Dr. Magdalena Lombardi Plasilova: So, we really advanced tremendously. AI is already used in imaging alongside mammograms, ultrasounds, breast MRI for screening and detection. It basically helps patients flagging suspicious areas on mammograms. The radiologist can prioritize, review the lesions, and diagnose cancer earlier. AI can help improve detection, both in mammograms, ultrasounds, MRIs, especially for patients, for example, with dense breasts where it's much harder to diagnose cancer much earlier. But it works together with radiologist's expertise, not alone.

Melanie Cole, MS: Okay. So, it's working alongside these things. Do you feel, in your own personal opinion, that it can help find cancers earlier than traditional methods? As we look at the different screening methods, there was normal mammography and then we got tomosynthesis, and then there's some other ways as well. Do you feel that it's helping really to find these things earlier?

Dr. Magdalena Lombardi Plasilova: Yes. AI actually can help detect cancer with traditional images in some instances with mammography, ultrasonography, and breast MRI earlier. It can detect cancers that can be missed by a human radiologist. So, catching things early can lead to earlier treatment and diagnosis. There have been many studies published, especially recently in Lancet, Mammography Screening with Artificial Intelligent, MASAI trial from Sweden, published in Lancet Digital Health that showed tremendous promises, using and reading images with radiologists and AI.

Melanie Cole, MS: Wow. So since you mentioned radiologists, is this an adjunct? Is it helping the radiologists or oncologists? How accurate is it compared to human radiologists reading our mammograms?

Dr. Magdalena Lombardi Plasilova: It can be definitely more accurate in terms of diagnosis. It can detect things as I mentioned earlier that a human can miss. And also, most importantly, it can help avoid us to do unnecessary biopsies and get results quicker.

Melanie Cole, MS: So, that's my next question. Do you feel that it will help to make less biopsies, quicker results, more accurate diagnoses, and as a result of all of those things, make the process-- which we already know is pretty stressful for women-- less stressful for patients in some way?

Dr. Magdalena Lombardi Plasilova: Definitely, Melanie. I'm sure you know somebody with breast cancer. This is a very stressful process. So, AI can help reduce wait times, waiting for pathology results, radiology results. I think, with use of AI, if we can reduce unnecessary biopsies for invasive procedures, it can be extremely beneficial for patients. And also with more precise diagnosis, improving diagnostic confidence, being able to tell patients that we have better diagnostic tools. I think it's all very beneficial.

Melanie Cole, MS: Yeah. I mean, we're all a little bit weary and wary about whether or not this is going to replace some of us. And you said we all know someone with breast cancer. Well, I told you off the air and most of my listeners know that I had it last year, so I know how stressful the situation can be. Do you feel that it could replace the screeners, radiologists in that situation for screening specifically at some point? Do you see that happening at all?

Dr. Magdalena Lombardi Plasilova: Don't worry, Melanie, you'll still have your doctor. Currently, AI is unlikely to replace doctors, but serves as a really powerful tool to augment and help their work. So, I think, AI can't replace doctors because it doesn't have, for example, empathy, critical judgment. So, AI is not going to be, or doctor only.

Melanie Cole, MS: So, it's helping with screening and diagnosis. Now, how about treatment decisions for breast cancer? There are so many tools in your toolbox now. We've got all sorts of therapies available, hormone therapies, targeted therapies, surgery, chemo, radiation, there's all kinds of things going on. So, how can it help the doctor in that shared decision-making with the patient in what's best for them?

Dr. Magdalena Lombardi Plasilova: As I mentioned, it utilizes advice from large data sets. So, it can help analyze genomic data to predict better treatment response, tumor behavior. So, that's where I think AI is extremely helpful for prediction of and helping guide the treatment based on the patient data, also clinical guidelines. AI models can help predict recurrence risks. I think it will be a good tool for monitoring. You know, there is a good role, for example, for variable devices and more prediction tools that can help oncologists tailor individual treatments. For example, some patients maybe do not need chemotherapies. With the help of AI, I think, that it'll be more personalized.

Melanie Cole, MS: Well, that's really where I see the future of medicine going in so many ways, Doctor, is personalized medicine. Are there concerns about bias in AI? Could it be less accurate for certain groups of women? Do we see that women with denser breasts or something like that high risk? Could it be more biased in that case? Are we really looking at the accuracy in any of those ways?

Dr. Magdalena Lombardi Plasilova: You are hitting the nail, Melanie. So, bias is one of the concerns for AI models. So yes, bias is a big concern. Because if we want to use the models, we have to really be able to trust the model. So, we know that AI models that are trained on data sets that like diversity, for example, the population is not very diverse, may be less accurate and basically less accurate for underrepresented groups. So, for example, as you mentioned, for patients with dense breast tissue, for example, if we train AI on models with patients that have less breast density, it's going to be very hard to expect to have a good accuracy to detect cancers by AI in dense breasts. So, bias is a big issue.

Melanie Cole, MS: I mean, I think this is really just ever-evolving, and it's a pretty exciting time in your field. Dr. Plasilova, what would you like patients to know or ask about AI if their hospital or clinic uses it in their breast cancer screening or treatment? Will they even know if it's being used?

Dr. Magdalena Lombardi Plasilova: That's a good question. Because AI is not very widely used yet, but it's actually a rapidly evolving field. So, patients, for example, can ask just simple questionnaire, how is it used in the healthcare system, in the hospital? Is it used in Radiology? Is it used with radiologist? Is it used for treatment planning? Is the hospital using FDA-approved models? Or also, I think important question from patients should be: Is there a good human oversight? And what kind of AI tools and models are used? Are those robust models that have been tested very well to minimize bias as we discussed earlier? So, there are many questions. How patients can address this, I think, would start with Radiology and Pathology, for example.

Melanie Cole, MS: Those were really good tips you just gave us and questions to ask, because those robust models that you're mentioning are really where we see those results and the outcomes. What excites you, Doctor, the most about the future of AI in breast cancer care? What do you see happening? What do you want to see happening and what's so exciting for you?

Dr. Magdalena Lombardi Plasilova: It's very exciting to see how the technology is evolving very fast. I'm excited to see AI improve early cancer detection, basically speeding up the diagnosis, treatments, tailoring better treatments. Some hospitals have, for example, AI that are participating in tumor boards. Those are all exciting new trends. AI is used, as I mentioned in Radiology, Pathology, Radiation Oncology and prediction. So, I'm excited that AI can help in all areas in cancer care to really advance the cancer field.

And the most, at the end, I'm excited for is the multi-model AI that integrates models that combine different areas. ,For example, imaging, genomic, clinical data. So, that's going to be an exciting future to have more precise and very accurate and fast field. And as a breast surgeon, I'm excited about, of course, AI in the surgical field, robotics, and augmented reality.

Melanie Cole, MS: Wow. This is a fascinating topic. And I think you and I could really talk a lot more. So, I hope you'll join us again as things start to evolve. And as we've heard today, artificial intelligence is not about replacing doctors, but about giving them sharper tools to detect and treat breast cancer more effectively. And for patients, this could mean earlier diagnoses, more personalized care, and greater access to expert level treatment no matter where you live.

And while AI is still evolving, I want listeners to know this, its role in breast cancer care is already making a difference and the future looks so promising. Thank you so much Dr. Plasilova, for shedding light on this exciting area of medicine. And Weill Cornell Medicine continues to see our patients in person as well as through video visits, and you can be confident of the safety of your appointments at Weill Cornell Medicine.

That concludes today's episode of Back to Health. We'd like to invite our audience to download, subscribe, rate, and review back to Health on Apple Podcast, Spotify, iHeart, and Pandora. For more health tips, go to weillcornell.org and search podcasts. And parents, don't forget to check out our Kids Health Cast. We have so many great podcasts there as well. I'm Melanie Cole. Thanks so much for joining us today.

Promo: Listen now to On The Mind, the new podcast from Weill Cornell Medicine, discussing the latest hot topics in psychiatry, psychology, and mental health. Join Dr. Daniel Knoepflmacher as he explores recent research in cutting-edge clinical care with leading scientists and providers. Learn new initiatives and community wellness and how to process your mental health journey while exploring everything on the mind. Subscribe wherever you listen to podcasts.

disclaimer: All information contained in this podcast is intended for informational and educational purposes. The information is not intended nor suited to be a replacement or substitute for professional medical treatment or for professional medical advice relative to a specific medical question or condition. We urge you to always seek the advice of your physician or medical professional with respect to your medical condition or questions. Weill Cornell Medicine makes no warranty, guarantee, or representation as to the accuracy or sufficiency of the information featured in this podcast, and any reliance on such information is done at your own risk.

Participants may have consulting, equity, board membership, or other relationships with pharmaceutical, biotech, or device companies unrelated to their role in this podcast. No payments have been made by any company to endorse any treatments, devices, or procedures. And Weill Cornell Medicine does not endorse, approve, or recommend any product, service, or entity mentioned in this podcast.

Opinions expressed in this podcast are those of the speaker and do not represent the perspectives of Weill Cornell Medicine as an institution.