| Both sides previous revision
Previous revision
Next revision
|
Previous revision
|
start [2018/07/25 10:04]
nigam [Selected Talks] |
start [2025/09/06 10:46] (current)
nigam |
| We analyze multiple data types (EHR, Claims, Wearables, Weblogs, and Patient blogs), to answer clinical questions, generate insights, and build predictive models at the [[http://bmir.stanford.edu/ | Stanford Center for Biomedical Informatics Research]]. We use machine learning, text-mining, and prior knowledge in medical ontologies to enable the learning health system. | We are a group of doctors, engineers, informatics professionals and students focused on enabling better care using existing health data. We develop novel methods to learn from patient-level health data, answer clinical questions that enable better medical decisions at the point of care, and have an active effort to research safe, ethical, and cost-effective strategies for using predictive models to guide mitigating care actions. Our research group is part of the Department of Medicine at Stanford, the Clinical Excellence Research Center, and the Department of Biomedical Data Science. |
| |
| ---- | ===== About us ===== |
| {{youtube>Njphqhju5Fo?size=840x470}} | [[:lab_members|Lab members]], [[:jobs| Open positions]], [[:blogs-and-media| Blogs and media]], [[https://news.google.com/topics/CAAqJAgKIh5DQkFTRUFvS0wyMHZNR295WDJoell4SUNaVzRvQUFQAQ?ceid=US:en&oc=3 | News and Press]] \\ |
| ---- | |
| |
| ====== Group information ====== | ===== Research ===== |
| |
| Current Group: [[Lab members]] \\ | We analyze multiple types of health data (EHR, Claims, Wearables, Weblogs, and Patient blogs), in service of the learning health system ([[:examples_of_prior_work|see examples]]). The work can be grouped into three focus areas: |
| On Boarding: [[int:Onboarding|New Lab members]], [[onboarding_affiliates|For Collaborators]] \\ | |
| Internal (log in required): [[int:Lab information]], [[int:Lab communication]], [[int:Projects]], [[int:rotation_projects|Rotations]], [[archive:|Archived pages]] \\ | |
| Contact: [[Nigam Shah]] | |
| |
| ====== Teaching ====== | - We **develop methods** to analyze multiple datatypes for generating insights such as detecting skin adverse reactions by analyzing content in a [[https://jamanetwork.com/journals/jamaoncology/fullarticle/2673831|health social network]], enabling [[https://pubmed.ncbi.nlm.nih.gov/31583282/|medical device surveillance]], discovering drug adverse events [[http://www.ncbi.nlm.nih.gov/pubmed/23571773| from clinical notes]] using novel methods for [[https://hai.stanford.edu/news/agile-nlp-clinical-text-covid-19-and-beyond|processing textual documents]]. |
| | - We **answer clinical questions** using aggregate patient data at the [[http://stanmed.stanford.edu/2016winter/on-the-button.html|bedside]]. The [[:greenbutton| green button project]] established the viability of this idea and led to the creation of [[https://www.atroposhealth.com/| Atropos Health]]. |
| | - We **build predictive models** that allow taking mitigating actions, [[https://stanmed.stanford.edu/artificial-intelligence-puts-humanity-health-care/|keeping the human in the loop]]. Research on [[:rail| foundation models]] from our team is put into practice by the [[https://dsatshc.stanford.edu/| Data Science team at SHC]]. |
| |
| * [[BIOMEDIN215|BIOMEDIN 215 Data Driven Medicine]] Autumn quarter of each year | |
| * [[AIHC Bootcamp| AI in Healthcare Bootcamp]] | |
| |
| ====== Research ====== | ===== Teaching ===== |
| |
| ==== Answering clinical questions ==== | === On campus === |
| | * [[https://shahlab.stanford.edu/bmds215/|BMDS 215]], taught for the DBDS Graduate program is designed to prepare you to pose and answer meaningful clinical questions using routinely collected healthcare data. |
| | * [[https://navigator.stanford.edu/classes/1266/17310|CIM 213]], taught for the MCiM program explores how to use electronic health records (EHRs) and other patient data in conjunction with recent advances in artificial intelligence (AI) and evolving business models to improve healthcare. |
| |
| | === Online === |
| | * [[https://online.stanford.edu/programs/artificial-intelligence-healthcare | Artificial Intelligence in Healthcare]], which reviews the current and future applications of AI in healthcare with the goal of learning to bring AI technologies into the clinic safely and ethically. |
| | * [[https://online.stanford.edu/programs/applications-machine-learning-medicine-program|Applications of Machine Learning in Medicine Program]], where you work through interactive exercises and case studies, attend live webinars, receive ongoing feedback from the course team, and collaborate with your fellow learners to gain the real-world skills doing machine learning projects. |
| | * [[https://online.stanford.edu/programs/generative-ai-technology-business-and-society-program#program-courses | Generative AI: Technology, Business, and Society Program]], which covers technical fundamentals, business implications, and societal considerations, all with a focus on putting people first. |
| |
| * [[http://content.healthaffairs.org/content/33/7/1229.abstract | A ‘Green Button’ For Using Aggregate Patient Data At The Point Of Care]]: We suggest a “green button” function within EHRs for clinicians to use aggregate patient data for real time decision making at the [[http://stanmed.stanford.edu/2016winter/on-the-button.html | point of care]]. Check out the [[greenbutton|idea]], or our recently launched [[ http://tinyurl.com/inf-consult-2017 | Informatics Consult Service]] that puts this idea in action. | ===== Public Talks ===== |
| * [[http://www.ped-rheum.com/content/11/1/45| Profiling risk factors for chronic uveitis in juvenile idiopathic arthritis]]: We report a new association between allergic conditions and chronic uveitis. Covered in Stanford Medicine Mag -- [[http://stanmed.stanford.edu/2014spring/article9b.html | Great Medical Mines]] and in the [[http://online.wsj.com/news/articles/SB10001424052702304536104579557851593416622 | Wall Street Journal]] | |
| * [[http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0124653 | Proton Pump Inhibitor Usage and the Risk of Myocardial Infarction in the General Population]]: This study provides an example of how a combination of experimental studies and data-mining approaches can be applied to prioritize drug safety signals for further investigation.See coverage in [[http://www.npr.org/sections/health-shots/2015/06/11/413433350/data-dive-suggests-link-between-heartburn-drugs-and-heart-attacks |NPR]], [[http://www.washingtonpost.com/news/to-your-health/wp/2015/06/10/common-heartburn-medications-linked-to-greater-risk-of-heart-attack/ |Washington Post]], [[http://well.blogs.nytimes.com/2015/06/10/gastric-reflux-drugs-linked-to-heart-attacks/?_r=0 |NY Times]], [[http://www.forbes.com/sites/robertglatter/2015/06/11/common-acid-reflux-drugs-associated-with-increased-risk-for-heart-attacks/ | Forbes]], [[ http://ww2.kqed.org/stateofhealth/2015/06/10/stanford-big-data-study-links-common-heartburn-drugs-with-heart-attack-risk/|KQED]], [[http://www.foxnews.com/health/2015/06/10/study-links-common-heartburn-drug-to-increased-heart-attack-risk/ |Fox News]], and [[http://www.scientificamerican.com/article/new-software-and-genetic-analyses-aim-to-reduce-problems-with-multiple-drug-combinations/ | Scientific American]]. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26644522 | Androgen Deprivation Therapy and Future Alzheimer's Disease Risk]]: This study found an association between the use of ADT in the treatment of prostate cancer and an increased risk of Alzheimer's disease in a general population cohort. | |
| |
| ==== Insights from data ==== | <html> |
| | <iframe width="1098" height="685" src="https://www.youtube.com/embed/videoseries?si=yOLA_66g0eXvYZyB&list=PL2ZpYYiSL_sovq7B1-iGQbjwGgMAvyzKQ" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> |
| | </html> |
| |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/23571773| Pharmacovigilance using clinical notes]]: Uses textual clinical notes for detecting single drug–adverse event associations (AUC of 80.4%) and for detecting drug–drug interactions (AUC of 81.5%). Press in [[http://www.forbes.com/sites/zinamoukheiber/2013/04/10/mining-electronic-health-records-reveals-clues-of-harmful-drug-reactions/|Forbes]], [[http://gigaom.com/2013/04/10/stanford-team-shows-how-doctors-notes-can-spot-problem-drugs/|GigaOM]]. Our efforts were the focus of an [[http://www.nature.com/clpt/journal/v93/n6/full/clpt201360a.html| editorial commentary]] titled //Advancing the Science of Pharmacovigilance//. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26293444 | Information Seeking and Drug-Safety Alert Response by Health Care Professionals]]: The information-seeking behavior associated with online evidence sources can offer insight into the information needs of health professionals and enable large-scale medical surveillance. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26707631 | An unsupervised learning method to identify reference intervals from a clinical database]]: We show that it is possible to use laboratory results and coded diagnoses to learn laboratory test reference intervals from clinical data warehouses. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26988586 | Postmarket Surveillance of Point-of-Care Glucose Meters through Analysis of Electronic Medical Records]]: We show that it is possible to assess device accuracy using coincident testing of point-of-care and central laboratory blood glucose measurements in a large cohort of critically ill patients. [[http://www.clinchem.org/content/62/5/668.extract | Editorial]] in Clinical Chemistry. | |
| |
| ==== Predictive Modeling ==== | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/24988898|Predicting Diagnoses of Depression]]: We developed a model that uses electronic medical record (EMR) data for predicting the diagnosis of depression up to 12 months before the diagnosis of depression. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26606167 | Rapid identification of slow healing wounds]]: We demonstrate that it is possible to build a model for identifying delayed healing wounds with an Area Under the Curve (AUC) of 0.842 that works across all wound types. | |
| * [[http://www.ncbi.nlm.nih.gov/pubmed/26483171 | Implications of non-stationarity on predictive modeling using EHRs]]: Under the non-stationarity in the underlying dataset, the performance advantage of complex methods such as stacking relative to the best simple classifier disappears. Ignoring non-stationarity can thus lead to sub-optimal model selection in predictive modeling tasks. | |
| * [[http://bmjopen.bmj.com/cgi/content/full/bmjopen-2016-011580?ijkey=oCxNIjOhCzOdmR8&keytype=ref | Predicting patient ‘cost blooms’]]: We develop models that identify new entrants to the upper decile of per capita healthcare expenditures in the next year. | |
| |
| ====== Selected Talks ===== | |
| {{youtube>xW3drA3ijRc?small | Building a Machine Learning Healthcare System, at XLDB 2018}} Building a Machine Learning Healthcare System, at XLDB 2018 | |
| ---- | |
| |
| {{youtube>2ERCBBQOMlg?small&start=460 | Performing an Informatics Consult, Grand rounds in Medicine at Stanford, Feb 1 2017}} Performing an Informatics Consult, Grand rounds in Medicine at Stanford, Feb 1 2017 | |
| ---- | |
| |
| [[Other talks]] | |
| ====== Seminars on campus ===== | |
| |
| [[Seminars]] | |
