Global AI and Data Science

Global AI & Data Science

Train, tune and distribute models with generative AI and machine learning capabilities

 View Only
Expand all | Collapse all

How can Artificial Intelligence help Healthcare - Health and AI

  • 1.  How can Artificial Intelligence help Healthcare - Health and AI

    Posted Mon February 10, 2020 11:29 AM
    One of the most noble application of AI is health, in my opinion. That's what brought me to change are in the IT world.

    I have already seen several amazing projects out there. But i would like you to "hear" from you all. Just use your imagination and tell your possible ideas when it comes to AI and Health. 


    ------------------------------
    Alisson Kalel
    Nutrition and Fitness Expert
    ------------------------------



    #DataandAILearning
    #AIandDSSkills
    #AIandDSSkills


  • 2.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Wed February 12, 2020 11:13 AM
    Hi,

    In X-Rays for example and by doing the administrative task of a nurse like filling forms and letting her spend more of her time on the  human side of her work

    Thanks

    ------------------------------
    Marcelo Silva
    ------------------------------



  • 3.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Wed February 12, 2020 02:15 PM
    Advancements Watson Healthcare has already made is incredible!  The concepts currently being researched are mindblowing!  As IBM leads the AI revolution, I anticipate and envision AI being able to:

    1. Instantly interpret imaging tests with immediate results being provided to doctors and patients.
    2. Evaluate multiple imaging and blood tests and correlate both current and potential concerns/risks.
    3. Reduce healthcare costs by providing doctors and patients with a highly encrypted/secure patient health portfolio that contains all current and historical health records.  This will eliminate redundant tests while giving practitioners access to tests hey may have skipped due to concerns about insurance coverage.
    4. Analyzing BILLIONS of research documents, trials, studies, health records, tests, and results in order to formulate both the most accurate diagnosis and the treatment with the greatest probability of success and the least harm and risk to the patient.
    New Brain-Twister
    Currently, drastic improvement to healthcare (globally) is at the top of the priority list.  What will take its place once we solve the healthcare goals and we are all living to the ripe age 150?

    ------------------------------
    David W. Kovalcik
    ------------------------------



  • 4.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Thu February 13, 2020 01:51 PM
    AI indeed is changing the viewpoint of many in the field of  Health Care 
    just a brief of a project which I am trying to build called Skin Buddy 

    a mobile application where people can use their phone camera to scan the area of the infected reason of the body and can get a detailed report on what it is and possible solutions

    ------------------------------
    shantam sultania
    ------------------------------



  • 5.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Mon February 17, 2020 04:45 PM
    Shantam, you're working in a really important area. This has been an active area of research for over 3 decades and has only increased in interest with the proliferation of smartphones. I recommend you search on arXiv for the most recent papers to see what others are doing in the field and on Google Scholar for some of the survey papers that discuss the current crop of apps doing similar things. 

    Because of the difficulty and expense of getting FDA approval for such an app, an approach which is gaining in popularity is Content Based Image Retrieval (CBIR) which simply shows users similar images with their diagnoses and letting the user decide what they might have and what course of action to take. This avoids making any kind of recommendation which is a trigger for regulatory oversight.

    If you haven't used it yet, you should look at the ISIC Archive for a dataset of over 23K labeled images. Granted, these are demoscopy images and not "clinical" images (the kind you would get with a smartphone) but it's a fantastic way to practice applying machine learning ideas to skin images.

    All the best...

    ------------------------------
    Tracy Petrie
    ------------------------------



  • 6.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Mon February 17, 2020 04:46 PM
    Hi Shantam,
    You're working on an important area but also one that is fairly crowded with other researchers. I highly recommend using arXiv to download papers on the subject and to look at the many similar apps that have already been published. Also, as you're working to understand how ML might work, checkout the ISIC Archive of dermoscopy images as a starting place if you haven't already. 

    One of the real problems in this area is the lack of data. This is especially true of disparate skin types. Most images are skin types I - II, with some III (i.e. lighter skinned). Another problem, which has been true for over 30 years of work in this area, is trying to answer the wrong problem. People often try to detect melanoma (the deadly skin cancer) vs. "benign nevi" but in the real world, people will take pictures of rashes, pimples, etc. If the classifier hasn't seen those during training, there is a good chance it will misclassify them which has potentially grave consequences both for the person and for the increased burden on the healthcare industry.

    Finally, look into what you'll need to do regarding FDA approval. Any kind of medical recommendation will require some expensive and time-consuming studies to prove effectiveness, etc. One approach that is gaining popularity is to simply show users "similar" images that have known (pathology-confirmed) outcomes and letting them decide what course of action to take.

    Good luck with your efforts,
    --Tracy

    ------------------------------
    Tracy Petrie
    ------------------------------



  • 7.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Thu March 12, 2020 02:08 PM
    There are many different ways that that AI can improve healthcare. As mentioned before, the use of X-rays has definitely come a long way. Another example of AI improving healthcare, is the fact that we now have technology that can detect AFib. Take your Apple Watch for example. A technology that is worn on your wrist now has the capability of letting you know if you have AFib, so that you are able to get a step ahead of it, and seek doctors, treatments, remedies, etc.

    ------------------------------
    Thomas Enriquez
    ------------------------------



  • 8.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Thu March 12, 2020 04:46 PM
    Just thinking out loud, I recently wanted to encourage someone who works with big company that creates creating artificial teeth to consider implementing some AI.  I was told that a lot of the work is labour intensive and that the production process also involved designing teeth using CAD. I was thinking that a lot of the design guidelines could be predefined and automated, but I wasn't sure what tools to encourage. Any suggestions?

    ------------------------------
    Lynn Petros
    ------------------------------



  • 9.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Wed January 08, 2025 03:53 PM
      |   view attached

    AI is helping in the medical field in many different ways, 0ne area where AI has helped some 35,000,00 patients has been in the treatment of epilepsy that affects 50,000,000 people, AI is being used in several different treatment options such as Vagal Nerve stimulation, ( a pacemaker type of unit that emits electronic signals as need to the human brain). Ictal and Inter-Ictal Recordings via Intracranial EEG & Stereotactic EEG are also used thru the placement of wires attached to the skull whereas AI reads neurol impulse's sending data read by AI with the physician making final treatment. It is also used in Seizure onset zone location (where the seizure develops) and for prediction and detection of seizures. There are wearable devices in addition all controlled thru AI.

    Machine learning plays a huge role in the fact that not every patient experiences the same type of seizure.

     Machine learning helps epilepsy patients by predicting future epilepsy, detecting and monitoring seizures, differentiating epilepsy from mimics, and improving neuroanatomic localization1Researchers have applied advanced machine learning technology to provide more accurate and rapid diagnoses2ML algorithms, particularly deep neural networks, are used to interpret electroencephalograms (EEG), neuroimaging, wearable data, and seizure videos3There have been developments in ML-based algorithms for early prediction of seizures using EEG signals4.

    For Full Review see: Artificial Intelligence and Computational Approaches for Epilepsy - PMC      

    #AIand health         #AIand Epilepsy    

    Information was found thru research, and thru personal communication with Dr. John C Jones, U.W. Madison Hospital & Professor at U.W Wis. Madison



    ------------------------------
    Thomas Mertens
    Dead End Fence
    Operator
    Medford, Wi U.S.A (Summer)
    Florida (Winter) U.S.A.
    ------------------------------

    Attachment(s)



  • 10.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Wed January 08, 2025 04:08 PM
    Very good. Now the question is how AI can become evil for humanity. Thats the question, cause the potential are unimagible and its almost impossible to grasp the scope of it and its repercussion.

    Marcelo Silva,
    Master Business Administrator





  • 11.  RE: How can Artificial Intelligence help Healthcare - Health and AI

    Posted Thu January 09, 2025 09:57 AM

    When everything is AI inclusive of healthcare are we going to lose the human touch of talking with someone vs a computer or a robot? Replication of the human touch AI can not duplicate. If AI is trained on biased data , you will get unwanted results which could lead to discriminatory  treatment of different age, race sex, etc... of individuals

    Security of AI is a major concern and should be prioritized.



    ------------------------------
    Thomas Mertens
    Medford, Wi U.S.A (Summer)
    Florida (Winter) U.S.A.
    ------------------------------