Description
March 5, 2026
Amyloidosis Workup: What Pathologists Need to Know
Pathology Made Simple: Chat GPT’s Summarization of Pathology Reports
AI in Healthcare Diagnostics: Promises, Pitfalls, and the Path Forward
Hosted on Ausha. See ausha.co/privacy-policy for more information.
Transcription
Cutting through questions about amyloid testing, CAP experts highlight key recommendations. Plus, new research puts CHAT-GPT through its paces summarizing pathology reports. Can the tool do it in one sentence? These stories and more next. This is the Path News Network Daily Edition from the College of American Pathologists. I'm Elizabeth McMahon. It's Thursday, March 5th. Amyloid accumulation is often associated with progressive diseases such as Alzheimer's and Parkinson's, but pathologists have long raised concerns that the techniques and criteria for identifying and classifying these abnormal protein deposits aren't standardized. Now, new CAP evidence-based guidelines address those worries with recommendations and good testing practices. In a recent episode of the CAPcast, Dr. Dylan Miller a general surgical pathologist who served on the guidelines expert panel, said the recommendations offered labs some flexibility in their use of amyloid laboratory stains.
The recommendation does, since it is evidence-based and since most of the evidence is with Congo Red, does recommend Congo Red, but does allow labs the opportunity to sort of validate whatever other stains they want to use as long as it shows an equivalent performance to Congo Red.
The recommendations also highlight the growing role of mass spectrometry in amyloid typing and addresses some of the sensitivity and specificity issues with other amyloid typing methods.
We definitely acknowledge the expense and the turnaround time delays that are involved with mass spectrometry fibrotyping. I think that it still is the best test. And so it is the recommended method.
You can find the formal recommendations and the full interview with Dr. Miller and Dr. Billy Fife-Kirchner at the CAP's Protocols and Guidelines page. With so many video streaming services at your fingertips, there's only one dedicated exclusively to your pathology practice. The members-only CAP on Demand is always expanding, and this spring, you'll find valuable new resources on topics that matter to you. In one newly released video, Dr. Kimberly Allison, Director of Breast Pathology at Stanford Medicine, discusses the challenges of evolving HER2 breast cancer guidelines for pathologists.
The HER2 guidelines before this were all designed to determine what the most predictive threshold was for response to treatment. And now we're in kind of different waters where there is not necessarily evidence of a new predictive threshold, but... you know, this is clinically relevant now, and how do we message this and report this?
You can find new releases like this one, plus videos on choroid plexus tumors, megaloblastic changes in bone marrow, and other topics when you log on to CAP on Demand. For busy clinicians, navigating lengthy pathology reports can be time-consuming and challenging. So wouldn't it be nice if ChatGPT could generate an accurate summary in one sentence? A new study in the March Archives of Pathology and Laboratory Medicine challenged the AI tool to do just that, and the results are promising. The research team from the Institute of Pathology at Sheba Medical Center in Israel evaluated 120 pathology reports covering four cancer types and compared the untrained chat GPT tools single-sentence summary to those of certified pathologists. Using a scale from 1 to 10, they scored ChatGPT on its summary of critical information, the mistakes it made, the essential information it included, and its readability. Though ChatGPT scored several 8s and 9s, the researchers noted it made errors and had limitations in handling complex cases. The authors also warned that using chat GPT. in medical settings raises ethical and legal questions. Speaking of legal and ethical questions, the International Graduate Business School, INSEAD, this week offers a deep dive into how AI can go right and wrong in healthcare diagnostics. The analysis by co-authors from INSEAD's Healthcare Ethics and Compliance Programs acknowledges the measurable impact AI has had on early diagnosis and treatment, but flags some limitations. In pathology, they pinpoint AI's vulnerability. to misinterpreting tissue contamination, among other challenges. They also warn that many AI models are trained on historical medical data, which may reflect long-standing inequalities in research and clinical practice. That's all for today's Daily Edition. Be sure to check the show notes for more information on today's stories. Watch your inbox for more news like this in the CAP's Advocacy Newsletter every Tuesday and our weekly edition newsletter. on Thursdays. We're back at 5 a.m. Eastern for another episode of The Daily Edition. I'm Elizabeth McMahon. Have a great day.
Description
March 5, 2026
Amyloidosis Workup: What Pathologists Need to Know
Pathology Made Simple: Chat GPT’s Summarization of Pathology Reports
AI in Healthcare Diagnostics: Promises, Pitfalls, and the Path Forward
Hosted on Ausha. See ausha.co/privacy-policy for more information.
Transcription
Cutting through questions about amyloid testing, CAP experts highlight key recommendations. Plus, new research puts CHAT-GPT through its paces summarizing pathology reports. Can the tool do it in one sentence? These stories and more next. This is the Path News Network Daily Edition from the College of American Pathologists. I'm Elizabeth McMahon. It's Thursday, March 5th. Amyloid accumulation is often associated with progressive diseases such as Alzheimer's and Parkinson's, but pathologists have long raised concerns that the techniques and criteria for identifying and classifying these abnormal protein deposits aren't standardized. Now, new CAP evidence-based guidelines address those worries with recommendations and good testing practices. In a recent episode of the CAPcast, Dr. Dylan Miller a general surgical pathologist who served on the guidelines expert panel, said the recommendations offered labs some flexibility in their use of amyloid laboratory stains.
The recommendation does, since it is evidence-based and since most of the evidence is with Congo Red, does recommend Congo Red, but does allow labs the opportunity to sort of validate whatever other stains they want to use as long as it shows an equivalent performance to Congo Red.
The recommendations also highlight the growing role of mass spectrometry in amyloid typing and addresses some of the sensitivity and specificity issues with other amyloid typing methods.
We definitely acknowledge the expense and the turnaround time delays that are involved with mass spectrometry fibrotyping. I think that it still is the best test. And so it is the recommended method.
You can find the formal recommendations and the full interview with Dr. Miller and Dr. Billy Fife-Kirchner at the CAP's Protocols and Guidelines page. With so many video streaming services at your fingertips, there's only one dedicated exclusively to your pathology practice. The members-only CAP on Demand is always expanding, and this spring, you'll find valuable new resources on topics that matter to you. In one newly released video, Dr. Kimberly Allison, Director of Breast Pathology at Stanford Medicine, discusses the challenges of evolving HER2 breast cancer guidelines for pathologists.
The HER2 guidelines before this were all designed to determine what the most predictive threshold was for response to treatment. And now we're in kind of different waters where there is not necessarily evidence of a new predictive threshold, but... you know, this is clinically relevant now, and how do we message this and report this?
You can find new releases like this one, plus videos on choroid plexus tumors, megaloblastic changes in bone marrow, and other topics when you log on to CAP on Demand. For busy clinicians, navigating lengthy pathology reports can be time-consuming and challenging. So wouldn't it be nice if ChatGPT could generate an accurate summary in one sentence? A new study in the March Archives of Pathology and Laboratory Medicine challenged the AI tool to do just that, and the results are promising. The research team from the Institute of Pathology at Sheba Medical Center in Israel evaluated 120 pathology reports covering four cancer types and compared the untrained chat GPT tools single-sentence summary to those of certified pathologists. Using a scale from 1 to 10, they scored ChatGPT on its summary of critical information, the mistakes it made, the essential information it included, and its readability. Though ChatGPT scored several 8s and 9s, the researchers noted it made errors and had limitations in handling complex cases. The authors also warned that using chat GPT. in medical settings raises ethical and legal questions. Speaking of legal and ethical questions, the International Graduate Business School, INSEAD, this week offers a deep dive into how AI can go right and wrong in healthcare diagnostics. The analysis by co-authors from INSEAD's Healthcare Ethics and Compliance Programs acknowledges the measurable impact AI has had on early diagnosis and treatment, but flags some limitations. In pathology, they pinpoint AI's vulnerability. to misinterpreting tissue contamination, among other challenges. They also warn that many AI models are trained on historical medical data, which may reflect long-standing inequalities in research and clinical practice. That's all for today's Daily Edition. Be sure to check the show notes for more information on today's stories. Watch your inbox for more news like this in the CAP's Advocacy Newsletter every Tuesday and our weekly edition newsletter. on Thursdays. We're back at 5 a.m. Eastern for another episode of The Daily Edition. I'm Elizabeth McMahon. Have a great day.
Share
Embed
You may also like
Description
March 5, 2026
Amyloidosis Workup: What Pathologists Need to Know
Pathology Made Simple: Chat GPT’s Summarization of Pathology Reports
AI in Healthcare Diagnostics: Promises, Pitfalls, and the Path Forward
Hosted on Ausha. See ausha.co/privacy-policy for more information.
Transcription
Cutting through questions about amyloid testing, CAP experts highlight key recommendations. Plus, new research puts CHAT-GPT through its paces summarizing pathology reports. Can the tool do it in one sentence? These stories and more next. This is the Path News Network Daily Edition from the College of American Pathologists. I'm Elizabeth McMahon. It's Thursday, March 5th. Amyloid accumulation is often associated with progressive diseases such as Alzheimer's and Parkinson's, but pathologists have long raised concerns that the techniques and criteria for identifying and classifying these abnormal protein deposits aren't standardized. Now, new CAP evidence-based guidelines address those worries with recommendations and good testing practices. In a recent episode of the CAPcast, Dr. Dylan Miller a general surgical pathologist who served on the guidelines expert panel, said the recommendations offered labs some flexibility in their use of amyloid laboratory stains.
The recommendation does, since it is evidence-based and since most of the evidence is with Congo Red, does recommend Congo Red, but does allow labs the opportunity to sort of validate whatever other stains they want to use as long as it shows an equivalent performance to Congo Red.
The recommendations also highlight the growing role of mass spectrometry in amyloid typing and addresses some of the sensitivity and specificity issues with other amyloid typing methods.
We definitely acknowledge the expense and the turnaround time delays that are involved with mass spectrometry fibrotyping. I think that it still is the best test. And so it is the recommended method.
You can find the formal recommendations and the full interview with Dr. Miller and Dr. Billy Fife-Kirchner at the CAP's Protocols and Guidelines page. With so many video streaming services at your fingertips, there's only one dedicated exclusively to your pathology practice. The members-only CAP on Demand is always expanding, and this spring, you'll find valuable new resources on topics that matter to you. In one newly released video, Dr. Kimberly Allison, Director of Breast Pathology at Stanford Medicine, discusses the challenges of evolving HER2 breast cancer guidelines for pathologists.
The HER2 guidelines before this were all designed to determine what the most predictive threshold was for response to treatment. And now we're in kind of different waters where there is not necessarily evidence of a new predictive threshold, but... you know, this is clinically relevant now, and how do we message this and report this?
You can find new releases like this one, plus videos on choroid plexus tumors, megaloblastic changes in bone marrow, and other topics when you log on to CAP on Demand. For busy clinicians, navigating lengthy pathology reports can be time-consuming and challenging. So wouldn't it be nice if ChatGPT could generate an accurate summary in one sentence? A new study in the March Archives of Pathology and Laboratory Medicine challenged the AI tool to do just that, and the results are promising. The research team from the Institute of Pathology at Sheba Medical Center in Israel evaluated 120 pathology reports covering four cancer types and compared the untrained chat GPT tools single-sentence summary to those of certified pathologists. Using a scale from 1 to 10, they scored ChatGPT on its summary of critical information, the mistakes it made, the essential information it included, and its readability. Though ChatGPT scored several 8s and 9s, the researchers noted it made errors and had limitations in handling complex cases. The authors also warned that using chat GPT. in medical settings raises ethical and legal questions. Speaking of legal and ethical questions, the International Graduate Business School, INSEAD, this week offers a deep dive into how AI can go right and wrong in healthcare diagnostics. The analysis by co-authors from INSEAD's Healthcare Ethics and Compliance Programs acknowledges the measurable impact AI has had on early diagnosis and treatment, but flags some limitations. In pathology, they pinpoint AI's vulnerability. to misinterpreting tissue contamination, among other challenges. They also warn that many AI models are trained on historical medical data, which may reflect long-standing inequalities in research and clinical practice. That's all for today's Daily Edition. Be sure to check the show notes for more information on today's stories. Watch your inbox for more news like this in the CAP's Advocacy Newsletter every Tuesday and our weekly edition newsletter. on Thursdays. We're back at 5 a.m. Eastern for another episode of The Daily Edition. I'm Elizabeth McMahon. Have a great day.
Description
March 5, 2026
Amyloidosis Workup: What Pathologists Need to Know
Pathology Made Simple: Chat GPT’s Summarization of Pathology Reports
AI in Healthcare Diagnostics: Promises, Pitfalls, and the Path Forward
Hosted on Ausha. See ausha.co/privacy-policy for more information.
Transcription
Cutting through questions about amyloid testing, CAP experts highlight key recommendations. Plus, new research puts CHAT-GPT through its paces summarizing pathology reports. Can the tool do it in one sentence? These stories and more next. This is the Path News Network Daily Edition from the College of American Pathologists. I'm Elizabeth McMahon. It's Thursday, March 5th. Amyloid accumulation is often associated with progressive diseases such as Alzheimer's and Parkinson's, but pathologists have long raised concerns that the techniques and criteria for identifying and classifying these abnormal protein deposits aren't standardized. Now, new CAP evidence-based guidelines address those worries with recommendations and good testing practices. In a recent episode of the CAPcast, Dr. Dylan Miller a general surgical pathologist who served on the guidelines expert panel, said the recommendations offered labs some flexibility in their use of amyloid laboratory stains.
The recommendation does, since it is evidence-based and since most of the evidence is with Congo Red, does recommend Congo Red, but does allow labs the opportunity to sort of validate whatever other stains they want to use as long as it shows an equivalent performance to Congo Red.
The recommendations also highlight the growing role of mass spectrometry in amyloid typing and addresses some of the sensitivity and specificity issues with other amyloid typing methods.
We definitely acknowledge the expense and the turnaround time delays that are involved with mass spectrometry fibrotyping. I think that it still is the best test. And so it is the recommended method.
You can find the formal recommendations and the full interview with Dr. Miller and Dr. Billy Fife-Kirchner at the CAP's Protocols and Guidelines page. With so many video streaming services at your fingertips, there's only one dedicated exclusively to your pathology practice. The members-only CAP on Demand is always expanding, and this spring, you'll find valuable new resources on topics that matter to you. In one newly released video, Dr. Kimberly Allison, Director of Breast Pathology at Stanford Medicine, discusses the challenges of evolving HER2 breast cancer guidelines for pathologists.
The HER2 guidelines before this were all designed to determine what the most predictive threshold was for response to treatment. And now we're in kind of different waters where there is not necessarily evidence of a new predictive threshold, but... you know, this is clinically relevant now, and how do we message this and report this?
You can find new releases like this one, plus videos on choroid plexus tumors, megaloblastic changes in bone marrow, and other topics when you log on to CAP on Demand. For busy clinicians, navigating lengthy pathology reports can be time-consuming and challenging. So wouldn't it be nice if ChatGPT could generate an accurate summary in one sentence? A new study in the March Archives of Pathology and Laboratory Medicine challenged the AI tool to do just that, and the results are promising. The research team from the Institute of Pathology at Sheba Medical Center in Israel evaluated 120 pathology reports covering four cancer types and compared the untrained chat GPT tools single-sentence summary to those of certified pathologists. Using a scale from 1 to 10, they scored ChatGPT on its summary of critical information, the mistakes it made, the essential information it included, and its readability. Though ChatGPT scored several 8s and 9s, the researchers noted it made errors and had limitations in handling complex cases. The authors also warned that using chat GPT. in medical settings raises ethical and legal questions. Speaking of legal and ethical questions, the International Graduate Business School, INSEAD, this week offers a deep dive into how AI can go right and wrong in healthcare diagnostics. The analysis by co-authors from INSEAD's Healthcare Ethics and Compliance Programs acknowledges the measurable impact AI has had on early diagnosis and treatment, but flags some limitations. In pathology, they pinpoint AI's vulnerability. to misinterpreting tissue contamination, among other challenges. They also warn that many AI models are trained on historical medical data, which may reflect long-standing inequalities in research and clinical practice. That's all for today's Daily Edition. Be sure to check the show notes for more information on today's stories. Watch your inbox for more news like this in the CAP's Advocacy Newsletter every Tuesday and our weekly edition newsletter. on Thursdays. We're back at 5 a.m. Eastern for another episode of The Daily Edition. I'm Elizabeth McMahon. Have a great day.
Share
Embed
You may also like