A.I. vs. Pathologists: Survival of the Fittest | Sahir Ali | TEDxSugarLand
Artificial Intelligence and it’s promise in predicting cancer outcome: every patient deserves their own equation. Dr. Sahirzeeshan Ali is a research scientist at the Center for Computation Imaging and Personalized Medicine (CCIPD) at Case Western Reserve Medical University and Seidman Cancer Center. Dr. Ali received a bachelor’s and master’s degrees in Electrical and Computer Engineering from Rutgers University (2009 & 2011) and a Ph.D in Biomedical Engineering from Case Western Reserve University. He also was the recipient of a Prostate Cancer Research Grant from the Department of Defense in 2014.
Dr. Ali’s research interest lies in developing image analysis, statistical pattern recognition, machine learning and artificial intelligence tools to computationally interrogate biomedical image data of digital pathology tissue images. The tools can be used to predict disease progression and provide a score to clinicians on the aggressiveness of a patient’s disease, such as breast cancer and prostate cancer, which can in turn help physicians decide on appropriate treatment option.
Dr. Ali has written more than 30 peer-reviewed journal, conference and abstract publications, appearing in journals such as Nature Scientific Reports, American Journal of Surgical Pathology, the Annual Review of Biomedical Engineering, Medical Image Analysis, IEEE Transactions on Medical Imaging. This research work has also culminated in various commercialized patents.
In addition, Dr. Ali has consulted with hedge funds and fortune 100 companies as a Salesforce architect and machine learning expert. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at www.ted.com/tedx
Пікірлер: 31
Beautiful talk and even more beautiful data
Excellent presentation...well delivered...
Great and impactful talk about the evolution about personalized medicine and the impact of technology such as AI.
Excellent
excellent. I have a question. But can we use it to predict if a pneumonia person is going to be aggressive or not? I have a final year project on that
this is super coo!
13:35: the system did pick one stromal cell instead of a tumour cell to design the bottom left green pattern. only pathologist eye can still distinguish this, but overall incredibly interesting.
@alicantuncel197
3 жыл бұрын
"Only pathologist eye can still distinguish this" argument is nonsense. Software's accuracy at this point is irrelevant. It shows that the work can be done if enough parameters are added.
@frantishekressel3910
3 жыл бұрын
@@alicantuncel197 Since Pathologists are the one to define the ground truth when training, this has massive sense.
@alicantuncel197
3 жыл бұрын
@@frantishekressel3910 The video has nothing to do with pathologists' work. That video is not about pathologists' contribution. It is about where AI is now, where it has the potential to reach in the future
@masterchief5603
2 жыл бұрын
@@alicantuncel197 you are approving something without a real evidence and full confidence from the folks knowing about this field massively way more than your imagination. Well go do a trip and do observe for _long time_ whatever goes into these fields. Don't talk before experiencing any of this yourself or until you have great certification from pathologist themselves.
Who takes responsibility when the AI goes wrong?
@hiryaa
6 жыл бұрын
I believe this is about a decision support system, and not a full replacement of humans. Just as optometrists now rely on automated readings (from machines) instead of traditional methods, these intelligent systems will enhance human shortcomings for better healthcare.
@chirag40625
5 жыл бұрын
Not pathologists.
@pangpingw
5 жыл бұрын
EKG machines also have its recommended diagnoses but Drs cannot trust that alone. Drs have to use their own judgments.
@medipoulet
5 жыл бұрын
People are alway looking how to get money from others. I hate patients like that.
@H3WSchmidt
3 жыл бұрын
Who takes responsibility if a doctor gets it wrong?
The title of this speech is way off base. It damages the credibility of the speaker because it illustrates his dramatic naivete. Pathologists need AI assistance to deal with the extreme volume of data, but AI needs Pathologists more to teach machine learning and pattern recognition machines to recognize cancer unequivocally. Fix the title of the talk.
@alicantuncel197
3 жыл бұрын
It does not illustrate any naivette. The title is actually extremely accurate. When software reaches enough accuracy it is going to replace pathologists, same is applicable to radiologists and dermatologists (on some level).
@masterchief5603
2 жыл бұрын
@@alicantuncel197 no evidence yet. All those models are irrelevant for self generating without the specialist. And in medicine there is no concept of "machine 100% accurate" which is actually what is mostly oversighted by folks.
@alicantuncel197
2 жыл бұрын
@@masterchief5603 nothing is "oversighted by folks" a ton evidence and currently working models exist, i suggest you to get your head out of where it is and do some proper literature search.
@alicantuncel197
2 жыл бұрын
@@masterchief5603 machines do not need to be 100% accurate although they come really close, they just need to get past doctors.
@alicantuncel197
2 жыл бұрын
@@masterchief5603 You are a joke. Years full of work and evidence exist and some dude with probably no basic understanding of research comes up talks like "no evidence yet". You have no idea what you are talking about and you are spreading misinformation.
But definitely it’s a group of pathologists who helped create this. What about their bias and setbacks. They will be impregnated into the AI as well and it will have same setbacks just in an exaggerated manner
I don’t agree with the title. Pathogists will benefit from having AI tools during S/O though
sometimes i wonder if it is not more like survival of the shrewdest, the meanest or the most physically strong