A.I. Shows Promise as a Physician Assistant - The New York Times

Bruno, Michael mbruno at PENNSTATEHEALTH.PSU.EDU
Tue Feb 12 15:04:40 UTC 2019


Exactly.  And there is a lot of bias, as was highlighted in this article:


[cid:image006.jpg at 01D4C2B8.C88BC080]

[https://cdn.technologyreview.com/i/images/screen-shot-2019-02-01-at-3.18.14-pm.png?sw=1180&cx=0&cy=0&cw=849&ch=478]
·
·
Intelligent Machines<https://www.technologyreview.com/topic/intelligent-machines/>
This is how AI bias really happens—and why it’s so hard to fix
Bias can creep in at many stages of the deep-learning process, and the standard practices in computer science aren’t designed to detect it.
·         by Karen Hao<https://www.technologyreview.com/profile/karen-hao/>  |   February 4, 2019


·

O


ver the past few months, we’ve documented how the vast majority<https://www.technologyreview.com/s/612404/is-this-ai-we-drew-you-a-flowchart-to-work-it-out/> of AI’s applications today are based on the category of algorithms known as deep learning, and how deep-learning algorithms<https://www.technologyreview.com/s/612437/what-is-machine-learning-we-drew-you-another-flowchart/> find patterns in data. We’ve also covered how these technologies affect people’s lives: how they can perpetuate injustice in hiring, retail, and security<https://www.technologyreview.com/s/612846/making-face-recognition-less-biased-doesnt-make-it-less-scary/> and may already be doing so in the criminal legal system<https://www.technologyreview.com/s/612775/algorithms-criminal-justice-ai/>.  But it’s not enough just to know that this bias exists. If we want to be able to fix it, we need to understand the mechanics of how it arises in the first place.



How AI bias happens

We often shorthand our explanation of AI bias by blaming it on biased training data. The reality is more nuanced: bias can creep in long before the data<https://dl.acm.org/citation.cfm> is collected as well as at many other stages<http://www.californialawreview.org/wp-content/uploads/2016/06/2Barocas-Selbst.pdf> of the deep-learning process. For the purposes of this discussion, we’ll focus on three key stages.



Framing the problem. The first thing computer scientists do when they create a deep-learning model is decide what they actually want it to achieve. A credit card company, for example, might want to predict a customer’s creditworthiness, but “creditworthiness” is a rather nebulous concept. In order to translate it into something that can be computed, the company must decide whether it wants to, say, maximize its profit margins or maximize the number of loans that get repaid. It could then define creditworthiness within the context of that goal. The problem is that “those decisions are made for various business reasons other than fairness or discrimination,” explains Solon Barocas, an assistant professor at Cornell University who specializes in fairness in machine learning. If the algorithm discovered that giving out subprime loans was an effective way to maximize profit, it would end up engaging in predatory behavior even if that wasn’t the company’s intention.



Collecting the data. There are two main ways that bias shows up in training data: either the data you collect is unrepresentative of reality, or it reflects existing prejudices. The first case might occur, for example, if a deep-learning algorithm is fed more photos of light-skinned faces than dark-skinned faces. The resulting face recognition system would inevitably be worse<https://www.technologyreview.com/s/612846/making-face-recognition-less-biased-doesnt-make-it-less-scary/> at recognizing darker-skinned faces. The second case is precisely what happened when Amazon discovered that its internal recruiting tool was dismissing female candidates<https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G>. Because it was trained on historical hiring decisions, which favored men over women, it learned to do the same.



Preparing the data. Finally, it is possible to introduce bias during the data preparation stage, which involves selecting which attributes you want the algorithm to consider. (This is not to be confused with the problem-framing stage. You can use the same attributes to train a model for very different goals or use very different attributes to train a model for the same goal.) In the case of modeling creditworthiness, an “attribute” could be the customer’s age, income, or number of paid-off loans. In the case of Amazon’s recruiting tool, an “attribute” could be the candidate’s gender, education level, or years of experience. This is what people often call the “art” of deep learning: choosing which attributes to consider or ignore can significantly influence your model’s prediction accuracy. But while its impact on accuracy is easy to measure, its impact on the model’s bias is not.



Why AI bias is hard to fix

Given that context, some of the challenges of mitigating bias may already be apparent to you. Here we highlight four main ones.

Unknown unknowns. The introduction of bias isn’t always obvious during a model’s construction because you may not realize the downstream impacts of your data and choices until much later. Once you do, it’s hard to retroactively identify where that bias came from and then figure out how to get rid of it. In Amazon’s case, when the engineers initially discovered that its tool was penalizing female candidates, they reprogrammed it to ignore explicitly gendered words like “women’s.” They soon discovered that the revised system was still picking up on implicitly gendered words<https://www.reuters.com/article/us-amazon-com-jobs-automation-insight/amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK08G>—verbs that were highly correlated with men over women, such as “executed” and “captured”—and using that to make its decisions.



Imperfect processes. First, many of the standard practices in deep learning are not designed with bias detection in mind. Deep-learning models are tested for performance before they are deployed, creating what would seem to be a perfect opportunity for catching bias. But in practice, testing usually looks like this: computer scientists randomly split their data before training into one group that’s actually used for training and another that’s reserved for validation once training is done. That means the data you use to test the performance of your model has the same biases as the data you used to train it. Thus, it will fail to flag skewed or prejudiced results.



Lack of social context. Similarly, the way in which computer scientists are taught to frame problems often isn’t compatible with the best way to think about social problems. For example, in a new paper<https://dl.acm.org/citation.cfm?id=3287598>, Andrew Selbst, a postdoc at the Data & Society Research Institute, identifies what he calls the “portability trap.” Within computer science, it is considered good practice to design a system that can be used for different tasks in different contexts. “But what that does is ignore a lot of social context,” says Selbst. “You can’t have a system designed in Utah and then applied in Kentucky directly because different communities have different versions of fairness. Or you can’t have a system that you apply for ‘fair’ criminal justice results then applied to employment. How we think about fairness in those contexts is just totally different.”



The definitions of fairness. It’s also not clear what the absence of bias should look like. This isn’t true just in computer science—this question has a long history of debate in philosophy, social science, and law. What’s different about computer science is that the concept of fairness has to be defined in mathematical terms, like balancing the false positive and false negative rates of a prediction system. But as researchers have discovered, there are many different mathematical definitions of fairness that are also mutually exclusive. Does fairness mean, for example, that the same proportion<https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing> of black and white individuals should get high risk assessment scores? Or that the same level of risk<https://www.washingtonpost.com/news/monkey-cage/wp/2016/10/17/can-an-algorithm-be-racist-our-analysis-is-more-cautious-than-propublicas/?utm_term=.2276d78de3c1> should result in the same score regardless of race? It’s impossible to fulfill both definitions at the same time (here’s<https://www.washingtonpost.com/news/monkey-cage/wp/2016/10/17/can-an-algorithm-be-racist-our-analysis-is-more-cautious-than-propublicas/?utm_term=.2276d78de3c1> a more in-depth look at why), so at some point you have to pick one. But whereas in other fields this decision is understood to be something that can change over time, the computer science field has a notion that it should be fixed. “By fixing the answer, you’re solving a problem that looks very different than how society tends to think about these issues,” says Selbst.



Where we go from here

If you’re reeling from our whirlwind tour of the full scope of the AI bias problem, so am I. But fortunately a strong contingent of AI researchers are working hard to address the problem. They’ve taken a variety of approaches: algorithms that help detect<https://arxiv.org/abs/1805.12002> and mitigate<http://aif360.mybluemix.net/> hidden biases within training data or that mitigate<https://www.technologyreview.com/the-download/612502/ai-has-a-culturally-biased-worldview-that-google-has-a-plan-to-change/> the biases<http://www.aies-conference.com/wp-content/papers/main/AIES-19_paper_220.pdf> learned by the model regardless of the data quality; processes<http://gendershades.org/overview.html> that hold companies accountable<http://www.aies-conference.com/wp-content/uploads/2019/01/AIES-19_paper_223.pdf> to the fairer outcomes and discussions<http://aif360.mybluemix.net/> that hash out the different definitions of fairness.  “‘Fixing’ discrimination in algorithmic systems is not something that can be solved easily,” says Selbst. “It’s a process ongoing, just like discrimination in any other aspect of society.”



This originally appeared in our AI newsletter The Algorithm.



https://www.technologyreview.com/s/612876/this-is-how-ai-bias-really-happensand-why-its-so-hard-to-fix/


From: Edward Hoffer [mailto:ehoffer at GMAIL.COM]
Sent: Tuesday, February 12, 2019 8:28 AM
To: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG
Subject: Re: [IMPROVEDX] A.I. Shows Promise as a Physician Assistant - The New York Times

Fascinating study.  The biggest problem with neural networks is their opacity - inability to explain in a comprehensible way why/how they reach their conclusions - which makes many reluctant to accept their conclusions.  The biggest problem with a "big data" approach is that one may be finding correlations rather than cause and effect, and correlation does not prove causation.  Only when these systems can explain their reasoning will they be widely accepted.
Ed
Edward P Hoffer MD
Co-creator, DXplain

On Mon, Feb 11, 2019 at 11:17 PM HM Epstein <hmepstein at gmail.com<mailto:hmepstein at gmail.com>> wrote:

I still believe that AI is there to help, not take over. But still an interesting article.
https://www.nytimes.com/2019/02/11/health/artificial-intelligence-medical-diagnosis.html<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nytimes.com_2019_02_11_health_artificial-2Dintelligence-2Dmedical-2Ddiagnosis.html&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=NneiGrfd_-1rLG-WeoP6kKrFMV6hwDbm2oHMeVnJQ0U&e=>

Best,
Helene

      [https://drive.google.com/uc?id=0BzqkjTYl6PKwNUhEU2NyZ2RsNW8&export=download]
A.I. Shows Promise as a Physician Assistant
Feb. 11, 2019

Doctors competed against A.I. computers to recognize illnesses on magnetic resonance images of a human brain during a competition in Beijing last year. The human doctors lost.Mark Schiefelbein/Associated Press
[https://static01.nyt.com/images/2019/02/12/science/12AIDIAGNOSIS/merlin_140515347_723de50e-b119-4cd4-8f57-56afc4f98d99-articleLarge.jpg?quality=75&auto=webp&disable=upscale]
Doctors competed against A.I. computers to recognize illnesses on magnetic resonance
images of a human brain during a competition in Beijing last year. The human doctors lost.


Each year, millions of Americans walk out of a doctor’s office with a misdiagnosis<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.sciencedaily.com_releases_2014_04_140416190948.htm&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=coJLyLfbRA0n5EtEfQ36gsByCvQYm43-vQq7XEGVac4&e=>. Physicians try to be systematic when identifying illness and disease, but bias creeps in. Alternatives are overlooked.  Now a group of researchers in the United States and China has tested a potential remedy for all-too-human frailties: artificial intelligence.

In a paper published on Monday in Nature Medicine, the scientists reported that they had built a system that automatically diagnoses common childhood conditions<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nature.com_articles_s41591-2D018-2D0335-2D9&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=4rgu2fxyWkQrljksOFViBLWgLTbmwVJX18H1OTx_K1g&e=> — from influenza to meningitis — after processing the patient’s symptoms, history, lab results and other clinical data.

The system was highly accurate, the researchers said, and one day may assist doctors in diagnosing complex or rare conditions.

Drawing on the records of nearly 600,000 Chinese patients who had visited a pediatric hospital over an 18-month period, the vast collection of data used to train this new system highlights an advantage for China in the worldwide race toward artificial intelligence.

Because its population is so large — and because its privacy norms put fewer restrictions on the sharing of digital data — it may be easier for Chinese companies and researchers to build and train the “deep learning” systems that are rapidly changing the trajectory of health care.

On Monday, President Trump signed an executive order<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nytimes.com_2019_02_11_business_ai-2Dartificial-2Dintelligence-2Dtrump.html-3Fmodule-3Dinline&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=s_M18PijwOkmPow_cN1zFtWL3pBcOndNFeN5OTnxN30&e=> meant to spur the development of A.I. across government, academia and industry in the United States. As part of this “American A.I. Initiative,” the administration will encourage federal agencies and universities to share data that can drive the development of automated systems.

Pooling health care data is a particularly difficult endeavor. Whereas researchers went to a single Chinese hospital for all the data they needed to develop their artificial-intelligence system, gathering such data from American facilities is rarely so straightforward.

“You have go to multiple places,” said Dr. George Shih, associate professor of clinical radiology at Weill Cornell Medical Center and co-founder of MD.ai, a company that helps researchers label data for A.I. services. “The equipment is never the same. You have to make sure the data is anonymized. Even if you get permission, it is a massive amount of work.”

After reshaping internet services, consumer devices and driverless cars in the early part of the decade, deep learning is moving rapidly into myriad areas of health care. Many organizations, including Google<https://urldefense.proofpoint.com/v2/url?u=https-3A__ai.googleblog.com_2018_05_deep-2Dlearning-2Dfor-2Delectronic-2Dhealth.html&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=xixMrdxJ587LueaCwEpk4ht6talAnZUOqQaNo52G0Ac&e=>, are developing and testing systems that analyze electronic health records in an effort to flag medical conditions such as osteoporosis, diabetes, hypertension and heart failure.

Similar technologies are being built to automatically detect signs of illness and disease in X-rays, M.R.I.s and eye scans.

The new system relies on a neural network<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nytimes.com_2018_03_06_technology_google-2Dartificial-2Dintelligence.html-3Fmodule-3Dinline&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=pha656hIAHSZ_RzcHJOJTLmydPTT8qt_wKvs7PErXtU&e=>, a breed of artificial intelligence that is accelerating the development of everything from health care to driverless cars<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nytimes.com_2018_01_04_technology_self-2Ddriving-2Dcars-2Daurora.html-3Fmodule-3Dinline&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=M_Hz0a7vEKnnZ1tkK58HFcy4H11uFBL--FDk8cS4YvU&e=> to military applications<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.nytimes.com_2018_02_20_technology_artificial-2Dintelligence-2Drisks.html-3Fmodule-3Dinline&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=Sn-KrLnn4m2YGhRL0HRwiEMWwG1ZYf7DAeSikMcbwgc&e=>. A neural network can learn tasks largely on its own by analyzing vast amounts of data.

Using the technology, Dr. Kang Zhang, chief of ophthalmic genetics at the University of California, San Diego, has built systems that can analyze eye scans for hemorrhages, lesions and other signs of diabetic blindness. Ideally, such systems would serve as a first line of defense, screening patients and pinpointing those who need further attention.

Now Dr. Zhang and his colleagues have created a system that can diagnose an even wider range of conditions by recognizing patterns in text, not just in medical images. This may augment what doctors can do on their own, he said.

“In some situations, physicians cannot consider all the possibilities,” he said. “This system can spot-check and make sure the physician didn’t miss anything.”

The experimental system analyzed the electronic medical records of nearly 600,000 patients at the Guangzhou Women and Children’s Medical Center in southern China, learning to associate common medical conditions with specific patient information gathered by doctors, nurses and other technicians.

First, a group of trained physicians annotated the hospital records, adding labels that identified information related to certain medical conditions. The system then analyzed the labeled data.

Then the neural network was given new information, including a patient’s symptoms as determined during a physical examination. Soon it was able to make connections on its own between written records and observed symptoms.

When tested on unlabeled data, the software could rival the performance of experienced physicians. It was more than 90 percent accurate at diagnosing asthma; the accuracy of physicians in the study ranged from 80 to 94 percent.

In diagnosing gastrointestinal disease, the system was 87 percent accurate, compared with the physicians’ accuracy of 82 to 90 percent.

Able to recognize patterns in data that humans could never identify on their own, neural networks can be enormously powerful in the right situation. But even experts have difficulty understanding why such networks make particular decisions and how they teach themselves.

As a result, extensive testing is needed to reassure both doctors and patients that these systems are reliable.

Experts said extensive clinical trials are now needed for Dr. Zhang’s system, given the difficulty of interpreting decisions made by neural networks.

“Medicine is a slow-moving field,” said Ben Shickel, a researcher at the University of Florida who specializes in the use of deep learning for health care. “No one is just going to deploy one of these techniques without rigorous testing that shows exactly what is going on.”

It could be years before deep-learning systems are deployed in emergency rooms and clinics. But some are closer to real-world use: Google is now running clinical trials of its eye-scan system at two hospitals in southern India.

Deep-learning diagnostic tools are more likely to flourish in countries outside the United States, Dr. Zhang said. Automated screening systems may be particularly useful in places where doctors are scarce, including in India and China.

The system built by Dr. Zhang and his colleagues benefited from the large scale of the data set gathered from the hospital in Guangzhou. Similar data sets from American hospitals are typically smaller, both because the average hospital is smaller and because regulations make it difficult to pool data from multiple facilities.

Dr. Zhang said he and his colleagues were careful to protect patients’ privacy in the new study. But he acknowledged that researchers in China may have an advantage when it comes to collecting and analyzing this kind of data.

“The sheer size of the population — the sheer size of the data — is a big difference,” he said.


________________________________

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG>

To unsubscribe from IMPROVEDX: click the following link:
http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1<https://urldefense.proofpoint.com/v2/url?u=http-3A__list.improvediagnosis.org_scripts_wa-2DIMPDIAG.exe-3FSUBED1-3DIMPROVEDX-26A-3D1&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=cJjpdW5-93UaYS_koIl_hqWUfswt-VmPLpJNJcXF4SM&e=> or send email to: IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG<mailto:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG>

Visit the searchable archives or adjust your subscription at: http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX<https://urldefense.proofpoint.com/v2/url?u=http-3A__list.improvediagnosis.org_scripts_wa-2DIMPDIAG.exe-3FINDEX&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=spdv6gvH7CKev454uDGkJa_IsJ4sv-uAxQnAVs2A8EI&e=>
Moderator:David Meyers, Board Member, Society for Improving Diagnosis in Medicine

To learn more about SIDM visit:
http://www.improvediagnosis.org/<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.improvediagnosis.org_&d=DwMFaQ&c=_FmMnDvUH5queZcSmOuBzHZMbp7E7EwtGwv5cxxnTj0&r=XZJky8Jx0OuETXcWpBMhx9j_wSYpSZPDVXdInJ5O9gQ&m=M4Uro1ASgsVLJljIOC7VNFztF3Rp8FGnxT9WEQBq-lw&s=c9ORCaJvVzvZQLxJTmVVKBB-xIpz30tPt8mSRAiU7Wo&e=>


To unsubscribe from the IMPROVEDX:
mail to:IMPROVEDX-SIGNOFF-REQUEST at LIST.IMPROVEDIAGNOSIS.ORG
or click the following link: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

Address messages to: IMPROVEDX at LIST.IMPROVEDIAGNOSIS.ORG

For additional information and subscription commands, visit:
http://www.lsoft.com/resources/faq.asp#4A

http://LIST.IMPROVEDIAGNOSIS.ORG/ (with your password)

Visit the searchable archives or adjust your subscription at:
http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?INDEX

Moderator: David Meyers, Board Member, Society to Improve Diagnosis in Medicine

To unsubscribe from the IMPROVEDX list, click the following link:<br>
<a href="http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1" target="_blank">http://list.improvediagnosis.org/scripts/wa-IMPDIAG.exe?SUBED1=IMPROVEDX&A=1</a>
</p>

HTML Version:
URL: <../attachments/20190212/6f36c9b3/attachment.html> ATTACHMENT:
Name: image001.jpg Type: image/jpeg Size: 5309 bytes Desc: image001.jpg URL: <../attachments/20190212/6f36c9b3/attachment.jpg> ATTACHMENT:
Name: image002.png Type: image/png Size: 205445 bytes Desc: image002.png URL: <../attachments/20190212/6f36c9b3/attachment.png> ATTACHMENT:
Name: AI for Diagnosis of Common Pediatric Illness.pdf Type: application/pdf Size: 2047717 bytes Desc: AI for Diagnosis of Common Pediatric Illness.pdf URL: <../attachments/20190212/6f36c9b3/attachment.pdf>


More information about the Test mailing list