BRAINX COMMUNITY LIVE !

May 2019 session.

At the first anniversary event of BrainX Community Live, Dr.Michael Burns made a presentation on “Practical Applications of Machine Learning and Natural Language Processing in Perioperative Medicine”. He is currently an anesthesiologist at the University of Michigan and clinical analytics researcher within the Multicenter Perioperative Outcomes Research (MPOG) group. He discussed current research and applications of Artificial Intelligence in perioperative patient care areas. These include applications of predictive analytics with neural networks and deep learning to guide diagnosis and treatment, especially useful in critically ill patients.

He also discussed his current work within MPOG with perioperative applications of machine learning models including classification models for Current Procedure Terminology (CPT) codes in medical procedural billing applications. Michael demonstrated that significant improvement can be achieved in classification of CPT codes using natural language processing and feature engineering. His work can be very impactful in increasing the speed, accuracy, and revenue capture compared to existing healthcare billing systems.

Links to his latest publications/presentations are below:

Mathur, P. and M. L. Burns (2019). "Artificial Intelligence in Critical Care." Int. Anesthesiology Clinics 57(2): 89-102. PMID: 30864993

"ProcedureView (ProView): Analyzing and Presenting Anesthesiology Case Data for Providers using Machine Learning" - Society for Technology in Anesthesia (2019)

"Clustering Anesthesiology Case Data for Machine Learning" - Machine Learning for Healthcare (2018)

 

 

BRAINX COMMUNITY LIVE !

April 2019 session.

Dr.Anant Madabhushi, was the presenter for the April 2019,BrainX Community live event.He is currently, Director of the Center for Computational Imaging and Personalized Diagnostics (CCIPD) and the F. Alex Nason Professor II in the Departments of Biomedical Engineering, Pathology, Radiology, Radiation Oncology, Urology, General Medical Sciences, and Electrical Engineering and Computer Science at Case Western Reserve University.

His presentation was on innovative use of machine learning techniques to aid with diagnosis, treatment and prognostication of cancer.He and his team have developed unique cost-effective solutions and pathbreaking techniques to provide less invasive, personalized and precise care to  cancer patients.They are currently researching application of these techniques to healthcare areas other than cancer including cardiology.

The key aspect of his presentation was demonstration of feature engineering as a key ML technique in the field of pathomics and radiomics to generate new and actionable knowledge.This new knowledge is enhancing  decision making for management of patients and supporting clinicians by providing them with actionable and precise information.

It is truly remarkable to see how Dr.Madabhushi is changing the management of cancer patients by providing actionable information in near real time and with significant accuracy.His orientation to patient care and clinician engagement is an important reason for this collaborative success.

He also discussed challenges associated with applications of machine learning in healthcare including access to data,quality of data,reproducibility of results,legal and ethical dilemmas.

Abstract to his presentation, links to his center and publications are available below.

 

 

Title - "Artificial Intelligence in Radiology and Pathology: Implications for Precision Medicine"

Abstract - Traditional biology generally looks at only a few aspects of an organism at a time and attempts to molecularly dissect diseases and study them part by part with the hope that the sum of knowledge of parts would help explain the operation of the whole. Rarely has this been a successful strategy to understand the causes and cures for complex diseases. The motivation for a systems based approach to disease understanding aims to understand how large numbers of interrelated health variables, gene expression profiling, its cellular architecture and microenvironment, as seen in its histological image features, its 3 dimensional tissue architecture and vascularization, as seen in dynamic contrast enhanced (DCE) MRI, and its metabolic features, as seen by Magnetic Resonance Spectroscopy (MRS) or Positron Emission Tomography (PET), result in emergence of definable phenotypes. At the Center for Computational Imaging and Personalized Diagnostics (CCIPD) at Case Western Reserve University, we have been developing computerized knowledge alignment, representation, and fusion tools for integrating and correlating heterogeneous biological data spanning different spatial and temporal scales, modalities, and functionalities. These tools include computerized feature analysis methods for extracting subvisual attributes for characterizing disease appearance and behavior on radiographic (radiomics) and digitized pathology images (pathomics). In this talk I will discuss the development work in CCIPD on new radiomic and pathomic approaches for capturing intra-tumoral heterogeneity and modeling tumor appearance. I will also focus my talk on how these radiomic and pathomic approaches can be applied to predicting disease outcome, recurrence, progression and response to therapy in the context of prostate, brain, rectal, oropharyngeal, and lung cancers. Additionally I will also discuss some recent work on looking at use of pathomics in the context of racial health disparity and creation of more precise and tailored prognostic and response prediction models.

Learn more about CCIPD:

http://engineering.case.edu/centers/ccipd/

CCIPD publications:

http://engineering.case.edu/centers/ccipd/publications

 

 

BRAINX COMMUNITY LIVE !

March 2019 session.

This session titled, “Hands on Tech:Learning AI”,focused on giving a primer on how to get started with machine learning in healthcare.

A step by step approach on training  in machine learning with a healthcare focus was provided by 

Piyush Mathur MD, FCCM , Founder of BrainX Community,Anesthesiologist/Intensivist, Quality Improvement Officer,Anesthesiology Institute, Cleveland Clinic.

Anirban Bhattacharyya MD, MPH , Associate Staff Physician,Respiratory Institute,Cleveland Clinic

Ghaith Habboub MD , Spine Fellow, Neurology Institute,Cleveland Clinic

Whether it is to write the codes or just to get familiar this approach is helpful in becoming bilingual and to foster collaborations between clinicians and machine learning experts.

A copy of the presentation is below:

BrainX Community Live! March 2019

 

 

 

 

BRAINX COMMUNITY LIVE !

January 2019 session.

A review of 2018 publications on ML applications in healthcare was presented by Piyush Mathur MD, FCCM.

Review and a copy of the presentation are available below:

PDF of 2018 Year in Review:Machine Learning in Healthcare.

BrainX Community Live January 2019 event

 

BRAINX COMMUNITY LIVE !

December 2018 session.

Presented by Feras Hatib, PhD, Director R&D, Edwards Lifesciences and Bart Geerts MD PhD MSc MBA, Consultant anaesthetist and researcher at AMC-Academic Medical Center,Netherland.They discussed  their research on prediction of hypotension through arterial waveform analysis using machine learning algorithm.All the way from selection of waveform variables to  2 million plus feature selection and engineering.Now paving the path for clinical trials and adoption.

Their research was recently published in the Anesthesiology journal’s October edition focused on Artificial Intelligence.Link and details of their publication are below:

http://anesthesiology.pubs.asahq.org/article.aspx?articleid=2685008

BRAINX COMMUNITY LIVE !

November 2018 session.

Dr.Ghaith Habboub,Spine fellow,Cleveland Clinic, discussed challenges with scaling and generalizing artificial intelligence(AI) applications in healthcare.

He pointed out that most of the effort currently goes into pre-processing and processing of data.

Machine learning applications are being created but integration into current workflow to make them actionable remains another challenge. Dr.Habboub presented some solutions to integrate both applications developed within EHR(electronic health record) and from outside of EHR.

Containerized API(application programming interface) appear to have certain advantages for embedding and scaling AI applications.

Presentation slides available via link below:

Embedding and Scaling AI Models in Healthcare Applications.11_3_2018

Follow his work on GitHb repository via link below:

https://github.com/rocketheat/Kubeadm_Rocketheat

Journal club on “Scalable and accurate deep learning with electronic health records”,generated interesting discussion on processing of the data using FHIR(Fast Healthcare Interoperability Resource).Use of multimodal data and automated feature engineering seems to be a strength of this study.

https://www.nature.com/articles/s41746-018-0029-1

For more information,contact Dr.Habboub directly at habboug@ccf.org

 

BRAINX COMMUNITY LIVE !

October 2018 session.

Dr. Aziz Nazha, Assistant Professor, Taussig Cancer Center, Cleveland Clinic started the October, BrainX Community live event by making a case for artificial intelligence applications  in medicine.He then went on to showcase some of the prognostic models in cancer that have been developed by his team and applications there of in personalized medicine.Newer models  using traditional recommender systems can be developed which use genomic biomarkers to predict response to therapy in a much more accurate manner than traditional methods.He also showed how neural networks can help in drug development and to target therapies.Integrating the work of pathomics and radiomics into cancer management along with electronic medical records data will hopefully build better treatment care paths in the future for more effective personalized care.

The Complexity of Interpreting Genomic Data in Patients with Primary and Secondary Acute Myeloid Leukemia (AML)
Aziz Nazha, Ahmad Zarzour, Tomas Radivoyevitch, Hetty E. Carraway, Jennifer S. Carew, Cassandra M Hirsch, Kassy E Kneen, Bartlomiej Przychodzen, Bhumika J. Patel, Michael Clemente, Srinivasa R. Sanikommu, Matt Kalaycio, Jaroslaw P. Maciejewski and Mikkael A. Sekeres
The Revised International Prognostic Scoring System "Molecular" (IPSS-Rm), a Validated and Dynamic Model in Treated Patients with Myelodysplastic Syndromes (MDS)
Aziz Nazha, Mayur Subhash Narkhede, Tomas Radivoyevitch, Matt Kalaycio, Bhumika J. Patel, Aaron T. Gerds, Sudipto Mukherjee, Michael Clemente, Cassandra M Hirsch, Anjali S. Advani, Bartlomiej Przychodzen, Hetty E. Carraway, Jaroslaw P. Maciejewski and Mikkael A. Sekeres
The efficacy of current prognostic models in predicting outcome of patients with myelodysplastic syndromes at the time of hypomethylating agent failure
Aziz Nazha, Rami S. Komrokji, Guillermo Garcia-Manero, John Barnard, Gail J. Roboz,David P. Steensma, Amy E. DeZern, Katrina Zell,Cassie Zimmerman, Najla Al Ali, Elias Jabbour,Molly D. Greenberg, Hagop M. Kantarjian, Jaroslaw P. Maciejewski, Alan F. List, and Mikkael A. Sekeres, On behalf of the MDS Clinical Research Consortium
Haematologica. 2016 Jun; 101(6): e224–e227.

BRAINX COMMUNITY LIVE !

September 2018 session.

Dr.Maheshwari, Director, Center for Perioperative Intelligence, shared his work on integrating data  intelligently into real time decision making to change perioperative outcomes thereby creating value.

He highlighted management of  important variables such as blood pressure in real time intra-operatively using workflow integrated advanced decision support.Features that have not been considered have the potential to be developed using machine learning and deep learning moving beyond simple models such as measurement of mean arterial pressure or  volume status alone.

Some of the references he cited are available below:

Relationship between Intraoperative Hypotension, Defined by Either Reduction from Baseline or Absolute Thresholds, and Acute Kidney and Myocardial Injury after Noncardiac Surgery: A Retrospective Cohort Analysis

Vafi Salmasi, M.D.; Kamal Maheshwari, M.D., M.P.H. ; Donsheng Yang, m.A.; Edward J. Mascha, Ph.D.;Asha Singh, M.D.; et al

http://anesthesiology.pubs.asahq.org/article.aspx?articleid=2579833

Association of Postoperative High-Sensitivity Troponin Levels With Myocardial Injury and 30-Day Mortality Among Patients Undergoing Noncardiac Surgery

Writing Committee for the VISION Study Investigators

https://jamanetwork.com/journals/jama/fullarticle/2620089

The association of hypotension during non-cardiac surgery, before and after skin incision, with postoperative acute kidney injury: a retrospective cohort analysis.

Maheshwari K, Turan A, Mao G, Yang D, Niazi AK, Agarwal D, Sessler DI, Kurz A.

https://onlinelibrary.wiley.com/doi/abs/10.1111/anae.14416

A Randomized Trial of Continuous Noninvasive Blood Pressure Monitoring During Noncardiac Surgery.

Maheshwari K, Khanna S, Bajracharya GR, Makarova N, Riter Q, Raza S, Cywinski JB, Argalious M, Kurz A, Sessler DI.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6072385/

The relationship between ICU hypotension and in-hospital mortality and morbidity in septic patients.

Maheshwari K, Nathanson BH, Munson SH, Khangulov V, Stevens M, Badani H, Khanna AK, Sessler DI.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6013508/


BRAINX
 COMMUNITY LIVE !

August 2018 session.

Our August session featured Pengtao Xie, PhD, who has recently completed his PhD from the Machine Learning Department at Carnegie Mellon University. Director of Data Services and Solutions and Research Scientist at Petuum,Inc.He presented his team’s work titled, “Deep Learning for Medical Information Extraction”.

We all learnt from the complexity of machine learning models and how they can be used to solve some of the real world electronic health record data problems with a high degree of accuracy.

Summary of the discussion from the 2 papers(in press) he discussed is below:

 

1) Named Entity Recognition (NER). 

Title: “Effective use of bidirectional language modeling for biomedical named entity recognition”

 

Background: There is an increased need for text mining in the biomedical field due to the rapid increase in the number of publications, scientific articles, reports, medical records, etc. that are available and readily accessible in electronic format. To transform unstructured collections of medical text into structured information and link them, information extraction systems must accurately identify different biomedical entities such as chemical ingredients, genes, proteins, medications, diseases, symptoms, etc. The task of identification and tagging of such entities in text as members of predefined categories such as diseases, chemicals, genes, etc. is referred to as NER. Designing an NER system with high precision and recall for the biomedical domain is a very challenging task due to the limited availability of high-quality labeled data and the linguistic variation of that data that includes ambiguous abbreviations, non-standardized descriptions, and lengthened names of entities. An NER system can be devised as a supervised ML task in which the training data consists of labels for each token in a text.

 

2) Relation Extraction (RE).

Title: “Relation Extraction of medical entities and attributes on Electronic Medical Records”.

 

This paper presents the findings of the development and testing of a deep learning model developed for medical relation extraction in clinical notes. The novel and effective deep learning approach automatically extracts relation for medical entities from EHRs. Specifically, a CNN-based model is used, which captures both salient syntactic feature and latent semantic feature from the text descriptions, despite their differences in language style. The model was evaluated on a real-patient dataset and achieved better performance than existing baselines on the tasks of extracting relations and deciding negations. It also shows significant potential in helping doctors in downstream tasks. 

Petuum has been named as a 2018 Technology Pioneer by the World Economic Forum with a significant focus and leadership in healthcare.

Link to Petuum's healthcare research and publications:

https://www.petuum.com/healthcare.html

 

BRAINX COMMUNITY LIVE !

July 2018 session.

July event featured Dr.Art Papier,who presented his 18 years experience in the field of diagnosis accuracy and use of visual diagnosis with the aid of machine learning to improve outcomes and decrease cost.

He showcased existing technology that his team has built which can be integrated into various workflows to aid decision making.

The talk was followed by a great discussion amongst engaged community members on various topics such as data access, patient privacy,regulatory/legal  issues,technological challenges and clinician adoption amongst others.

Link to Dr. Art Papier's July 2018 BrainX Community session video is attached below.

Note:It opens with Internet Explorer browser only.

http://webcast1.ccf.org/viewerportal/ccfe/video.vp?programId=esc_program:107577

A commentary by Dr. Papier recently published in the American Journal of Medicine July issue.

https://www.amjmed.com/article/S0002-9343(18)30094-9/pdf

Link to 1971 Larry Weed Internal Medicine Grand Rounds at Emory as discussed by Dr.Papier's during his presentation.

"It's incredible to see what Dr. Weed visioned 47 years ago, so much is relevant today……"

https://www.visualdx.com/company/larry-weed-1971-grand-rounds-at-emory-video

Learn more about CoreML, VisualDx.

http://www.businessinsider.com/visualdx-machine-learning-app-for-skin-diagnosis-ceo-interview-2017-11

Learn more about the Visual Dx through the link below:

https://www.visualdx.com

IMG_0024

IMG_0030

IMG_0032

IMG_0035

BRAINX COMMUNITY LIVE !

June 2018 Session

June Session featured Satish Viswanath,PhD. who gave a presentation on Machine Learning applications in Oncology and GastroIntestinal diseases.

His presentation focused on his teams application of combining Imaging data with pathology features to augment understanding of malignancy and it's treatment effect.

Literature review available via link below:

https://www.brainxai.org/learn/

Learn more about Dr.Viswanath and his teams work through link below:

http://engineering.case.edu/centers/ccipd/

IMG_1599

IMG_1601

BRAINX COMMUNITY LIVE !

Innaugral May 2018 session.

Inaugural session of BrainX Community's first live session held at Cleveland Clinic in May,2018.

Theme:Can AI transform healthcare?

IMG_0015
IMG_0012

Home Forums

    • Forum
    • Topics
    • Posts
    • Freshness
    • BEHAVIORAL SCIENCE
      This forum is to discuss applications of machine learning in the field of behavioral science.
    • 0
    • 0
    • No Topics

    • DERMATOPATHOLOGY
      This forum is to discuss applications of machine learning to advance the field of dermatopathology.
    • 0
    • 0
    • No Topics

    • IMAGING
      This forum is to discuss applications for Imaging in healthcare using Machine learning
    • 0
    • 0
    • No Topics

    • ONCOLOGY
      This forum is to discuss applications for Oncology using Machine Learning.
    • 0
    • 0
    • No Topics

    • PREDICTIVE ANALYTICS
      This forum is to discuss applications for predictive analytics using machine learning.
    • 0
    • 0
    • No Topics