Latest / Morning Medical with Dr. Chow / AI in Cardiovascular Care, Breast Cancer Disparities, and Nova Scotia Parking Policies

Transcript

AI in Cardiovascular Care, Breast Cancer Disparities, and Nova Scotia Parking Policies · Morning Medical with Dr. Chow

• 0:00
  Welcome to another episode of Morning Medical with Dr.
• 0:03
  Chow.
• 0:04
  I'm your host, Dr.
• 0:06
  Chi-Ming Chow.
• 0:07
  In today's episode, we'll be diving into some captivating health stories that are making
• 0:12
  headlines around the globe.
• 0:14
  We'll explore why there's hesitation about free hospital parking, how artificial intelligence
• 0:19
  is breaking the 'intellectual bottleneck' in healthcare, and the significant decline in
• 0:24
  breast cancer mortality among young women after 2010.
• 0:28
  Let's get started.
• 0:33
  First on the docket, let's talk about a new policy in Nova Scotia, Canada, where parking at
• 0:39
  all hospitals will be free for patients and visitors as of midnight.
• 0:44
  This was a promise made by Tim Houston's Progressive Conservative party during the
• 0:48
  election.
• 0:50
  While this might sound like a win for patients and their families, it hasn't come without some
• 0:54
  hesitation.
• 0:56
  Free hospital parking is seen as a way to alleviate some of the financial burdens that
• 1:01
  patients and families face during hospital visits.
• 1:04
  For many, the costs of parking can add up quickly, especially during prolonged hospital
• 1:09
  stays or frequent visits for ongoing treatments.
• 1:13
  By removing this cost, the government aims to make healthcare more accessible and reduce the
• 1:18
  stress associated with hospital visits.
• 1:21
  However, there are concerns about the potential drawbacks of this policy.
• 1:26
  Some worry that free parking might lead to overcrowding in hospital parking lots.
• 1:31
  This could make it difficult for patients and visitors to find parking spaces, potentially
• 1:36
  causing delays and additional stress.
• 1:39
  Additionally, there's the question of how hospitals will cover the lost revenue that
• 1:44
  parking fees previously provided.
• 1:47
  This income often helps fund hospital operations, and without it, there may be
• 1:53
  financial shortfalls that need to be addressed.
• 1:56
  Moreover, this move comes at a time when the healthcare system is already under financial
• 2:01
  strain.
• 2:03
  Critics argue that while the intention of free parking is good, the timing might not be ideal
• 2:09
  given the current challenges faced by hospitals.
• 2:13
  Balancing the need to support patients financially with the operational needs of
• 2:17
  healthcare facilities is a delicate task.
• 2:21
  In summary, while free hospital parking in Nova Scotia is a step towards reducing the financial
• 2:27
  burden on patients, it raises important questions about implementation and
• 2:31
  sustainability.
• 2:33
  As with any policy change, it's crucial to monitor its impact closely to ensure that it
• 2:38
  achieves its intended goals without unintended negative consequences.
• 2:43
  Now, let's delve into how artificial intelligence is helping to break what experts
• 2:47
  call the 'intellectual bottleneck' in healthcare.
• 2:51
  At the University of Texas Medical Branch, they've implemented a fascinating use of
• 2:56
  artificial intelligence that is transforming how they manage cardiovascular risk among their
• 3:01
  patients.
• 3:02
  Whenever a patient undergoes a computerized tomography scan at this facility, the images
• 3:08
  are automatically sent to the cardiology department, where artificial intelligence
• 3:13
  analyzes them and assigns a cardiac risk score.
• 3:17
  This process has been incredibly effective in flagging patients at high cardiovascular risk,
• 3:23
  even when the scan isn't specifically related to the heart.
• 3:27
  The beauty of this system is its simplicity and its wide-reaching potential.
• 3:32
  Artificial intelligence doesn't need to perform any superhuman feats to be valuable.
• 3:37
  Instead, it executes what are considered low-intellect tasks, but it does so at such a
• 3:42
  high volume that it unveils significant findings that might otherwise go unnoticed.
• 3:48
  According to Peter McCaffrey, the chief artificial intelligence officer at the
• 3:52
  University of Texas Medical Branch, "The data is just sitting out there." Artificial
• 3:58
  intelligence is enabling them to finally utilize these vast amounts of data that are
• 4:02
  often overlooked.
• 4:05
  One of the primary uses of artificial intelligence at the University of Texas Medical
• 4:09
  Branch is cardiac risk screening.
• 4:11
  The facility has trained models to scan for incidental coronary artery calcification, which
• 4:17
  is a strong predictor of cardiovascular risk.
• 4:20
  The goal is to identify patients who might be at risk of heart disease but show no obvious
• 4:25
  symptoms.
• 4:26
  Every computerized tomography scan, regardless of its original purpose, is analyzed by
• 4:31
  artificial intelligence to detect these calcifications.
• 4:35
  This is done using an image-based convolutional neural network that calculates an Agatston
• 4:40
  score, which reflects the accumulation of plaque in a patient's arteries.
• 4:46
  What’s particularly interesting is the way artificial intelligence categorizes patients
• 4:51
  based on their coronary artery calcification scores.
• 4:54
  Patients with a score of one hundred or more, who have no known history of cardiology visits
• 4:59
  or therapies, are automatically sent digital notifications, and their primary physicians are
• 5:05
  informed.
• 5:06
  For those with more severe scores, artificial intelligence prompts a phone call, ensuring
• 5:11
  they receive the necessary follow-up.
• 5:14
  The facility is even piloting tools to automate these calls.
• 5:19
  Beyond cardiovascular screening, artificial intelligence is also used in detecting strokes
• 5:24
  and pulmonary embolisms.
• 5:26
  Specialized algorithms have been trained to spot specific symptoms and alert care teams
• 5:31
  within seconds of imaging.
• 5:33
  This rapid notification system can significantly accelerate treatment, which is
• 5:39
  crucial in emergency situations where every minute counts.
• 5:43
  The artificial intelligence analyzes the computerized tomography scans for signs like
• 5:48
  obstructed blood flows or abrupt blood vessel cutoffs, doing so in mere seconds.
• 5:54
  Another fascinating aspect of their artificial intelligence implementation is its role in
• 5:59
  reducing errors like hallucinations and anchoring bias.
• 6:04
  The University of Texas Medical Branch has developed methods to ensure artificial
• 6:08
  intelligence models perform optimally, both before and after deployment.
• 6:13
  They compare artificial intelligence results with manual radiologist scores to calculate
• 6:18
  error rates and detect biases.
• 6:21
  They also use a peer learning technique where radiology exams are randomly assigned to
• 6:25
  different radiologists to assess performance and potential anchoring bias.
• 6:30
  Ultimately, the use of artificial intelligence at the University of Texas Medical Branch
• 6:36
  exemplifies how healthcare facilities can leverage technology to enhance patient care.
• 6:42
  By flagging what McCaffrey calls "basic stuff," artificial intelligence helps to overcome the
• 6:49
  intellectual bottleneck in healthcare.
• 6:52
  The sheer volume of data in healthcare is often too vast for human manpower alone to process.
• 6:59
  Artificial intelligence enables facilities to proactively find issues earlier, potentially
• 7:05
  leading to better patient outcomes.
• 7:08
  Breast cancer has long been a significant health concern, especially among young women.
• 7:14
  However, there's a glimmer of hope on the horizon.
• 7:18
  Researchers have discovered that breast cancer mortality rates have declined significantly
• 7:24
  among women aged 20 to 49 over the past decade.
• 7:28
  This finding was presented by Dr.
• 7:30
  Adetunji Toriola from the Siteman Cancer Center at Washington University School of Medicine
• 7:35
  during the American Association for Cancer Research annual meeting.
• 7:40
  Between 2010 and 2020, mortality rates for this age group dropped from 9.70 per 100,000 women
• 7:48
  to just 1.47 per 100,000.
• 7:52
  These declines were observed across all tumor subtypes and racial or ethnic groups,
• 7:58
  showcasing a broad improvement in outcomes.
• 8:01
  The steepest decline began around 2016, correlating with advancements in treatment
• 8:07
  options, the adoption of precision medicine, and better access to care and screening.
• 8:14
  Despite these positive trends, disparities still exist.
• 8:18
  For example, Black women, while experiencing significant reductions in mortality, still have
• 8:25
  higher incidence-based mortality rates compared to other groups.
• 8:29
  In 2010, the rate was 16.56 per 100,000, and although it decreased, it remained the highest
• 8:36
  at 3.41 per 100,000 in 2020.
• 8:41
  Comparatively, White women had the lowest rates, dropping from 9.18 to 1.16 per 100,000
• 8:48
  over the same period.
• 8:50
  The survival rates also tell an encouraging story but highlight areas needing improvement.
• 8:54
  Five-year relative survival rates exceed 88.7% for most groups, but Black women have a lower
• 9:02
  rate at 82.4%.
• 9:04
  This gap persists even at the 10-year mark.
• 9:08
  Dr.
• 9:08
  Toriola emphasized the need to address these disparities to ensure all women benefit equally
• 9:15
  from advancements in breast cancer treatment and care.
• 9:19
  Interestingly, the study also found that older women between 40 and 49 had better survival
• 9:25
  rates than younger women.
• 9:27
  This difference underscores the need for targeted screening and interventions,
• 9:31
  particularly for high-risk groups under 40.
• 9:34
  Molecular subtype analysis revealed that the luminal A subtype, generally considered less
• 9:40
  aggressive, had lower survival rates in younger women than expected, which was an unexpected
• 9:46
  finding that researchers are keen to explore further.
• 9:50
  Dr.
• 9:51
  Elaine R.
• 9:52
  Mardis, a moderator at the press briefing, found these results particularly intriguing,
• 9:58
  noting the unexpected crossover in survival rates between luminal A and luminal B subtypes
• 10:03
  in younger women.
• 10:05
  Dr.
• 10:06
  Toriola suggested that the biology of luminal A tumors in younger women might be different,
• 10:12
  potentially more aggressive or resistant to current therapies, especially since these
• 10:17
  tumors are HER2-negative.
• 10:20
  This study highlights the importance of continuous research to understand the
• 10:25
  underlying biological differences in breast cancer among various age groups and subtypes.
• 10:32
  It's crucial for developing more effective, personalized treatment strategies.
• 10:38
  As we move forward, ensuring equitable access to these advances and addressing the remaining
• 10:43
  disparities will be essential in continuing the progress against breast cancer.
• 10:49
  That's all for today's episode of Morning Medical with Dr.
• 10:53
  Chow.
• 10:54
  Thank you for joining me as we explored these important health stories.
• 10:58
  Remember, knowledge is power, especially when it comes to your health.
• 11:03
  Stay informed, stay healthy, and I'll see you next time.
• 11:07
  Goodbye!