Taipei Medical University (TMU) collaborates with Massachusetts Institute of Technology(MIT) to promote an integrative program designed to build local capacity and enhance locally sustained innovation. The program is deliberately structured to advance projects from idea to implementation.

MULTIDISCIPLINARY

This datathon brings together students, clinicians, data scientists and innovators to share ideas and bring theory into practice. The 3-day event allows participants to immediately apply concepts to actual projects, and work in teams to develop rapid prototypes and solve real problems. These projects may then be continued as research project or startup company.

IMPROVE HEALTHCARE

This Datathon aims to the further innovation in Risk Prediction, Disease Progression Prediction, Prognosis Prediction and Mortality Prediction to improving healthcare.

STEP1: Fill out the Registration Questionnaire .

STEP2: Pre-Order tickets on Accupass.

STEP3: After reviewing and decision is out, we'll send a confirmation email and approve your purchase.

※Note: We will only approve your purchase after the final list is out. Final list of confirmed participants will be announced after registration is closed. Final list will be decided by TMU-MIT Healthcare Datathon committee.

STEP4: Pay your ticket through Accupass.

STEP5: Get ready to be hero.

◆6/1-6/30 Early Birds
Non-Student: NT$ 1,600/person
Student: NT$ 400/person

◆7/1-8/31 Standards
Non-Student: NT$ 2,000/person
Student: NT$ 500/person

◆6/1-8/31 ELECT/UST Students: NT$ 500/person
Students from ELECT (Excellent Long-Established University Consortium of Taiwan) or UST (University System of Taipei) can receive NT$500 registration fee refund if they achieve perfect attendance.
To receive credit for the course Healthcare Analytics, please add to the roll on ELECT website between 8/14-8/27. For UST, hard copy application is required.
ELECT: SCU, CCU, SHU, TKU, MCU, FJU, USC, TMU, TMU, CYCU, FCU, PU
UST: NTPU, NTUT, TMU, NTOU

◆6/1-8/31 Physicians (By Invitation): NT$ 2,000/person
Please fill in your Receipt Title and Tax ID Number carefully on payment page for later reimbursement information.
If you are a physician without an invitation but would like to attend this event, please contact Howard (dd23245@tmu.edu.tw).

※Note: Receipt will be provided on the event day. For reimbursement needed, please fill in your Receipt Title and Tax ID Number carefully on payment page.

※Grouping method: The goal is to group participants from different fields into one team so it will be 4-5 members assigned by TMU in each team.

※Note: All participants are required to sign non-disclosure agreement, filming agreement and any other required agreements before the event start.

※Note: According to Income Tax Act, 10% of prize money over NT$20,000 is withheld for taxes. For non-resident winners, the withholding tax rate shall be in accordance with the associated regulations.

9/7(Sat) Pre-Datathon
09:00-09:30	Registration
09:30-09:35	Opening Remark
09:35-09:40	Introduction to Pre-Datathon
09:40-09:50	Introduction to MIMIC, eICU
10:00-11:00	Physicians Pitch (3 min/person)
11:00-12:00	Discussion

※Note:
-Physicians will take turns to pitch their clinical challenges and be the leads for our datathon teams.
-Participants are encouraged to discuss with the physicians and fill a form to list the clinical challenges in order of preference at the end of Pre-Datathon.
-Teams will be formed by the Datathon committee according to participants' preference and announced on 9/24.

9/27(Fri) Day 1
09:00-09:30	Registration
09:30-10:00	Opening Ceremony
10:00-10:30	Introduction to TMUxMIT Datathon/Mentors
10:30-10:50	Keynote Speech Speaker: Kenneth Paik Topic: Building Systems of Innovation
10:50-11:10	Tea Break
11:10-11:30	Keynote Speech Speaker: Jennifer Jordan Topic: The Intersection of AI & Ethics - Managing the Life Cycle of Predictive Applications to Reduce Risk
11:30-11:50	Keynote Speech Speaker: Wei-Hung Weng Topic: How to Conduct a Machine Learning Project for Clinical Medicine
11:50-13:00	Lunch
13:00-13:30	Workshop MIMIC III Speaker: Han Wang, National University of Singapore
13:30-14:00	Workshop The eICU Collaborative Research Database Speaker: Wei-Hung Weng, MIT EECS and Computer Science and AI Laboratory
14:00-18:00	Hacking
18:00-19:00	Dinner
19:00-20:00	1st Pre-Pitch (3 min Present + 3 min Q&A) Format Fixed
20:00-	Hacking

9/28 (Sat) Day 2
08:30-09:00	Breakfast/Registration
09:00-09:20	Keynote Speech Speaker: Rani Shifron Topic: AI in Healthcare: Enhancing the Efficacy and Efficiency of Interactions to the Rural World
09:20-09:40	Keynote Speech Speaker: Peggy Lai Topic: Time-Limited Trials for Critically Ill Cancer Patients
09:40-10:00	Keynote Speech Speaker: Mengling Feng Topic: When Machine Learning Meets Healthcare
10:00-12:00	Hacking
12:00-13:00	Lunch
13:00-15:00	Hacking
15:00-16:00	2nd Pre-Pitch (3 min Present + 3 min Q&A)
16:00-18:00	Hacking
18:00-19:00	Dinner
19:00-	Hacking

9/29 (Sun) Day 3
07:00-08:00	Breakfast/Registration
08:00-08:20	Mentors Meeting/Refresh on Evaluation Standard
08:20-10:50	Pitch (5 min Pitch + 3 min Q&A)
10:50-11:10	Evaluation + Questionnaire
11:10-11:25	Results Announcement + Final Comment
11:25-11:35	Award Ceremony
11:35-11:45	Closing Ceremony
11:45-12:00	Photo Time

※Note: Agenda may be subject to change.

※Note:
1.Access control will push between 00:00-06:00.
2.No sleeping bag will be provided, please bring your own if necessary.
3.Shower room will be available at certain time.

※Note: Participants are required to attend for both Pre-Datathon and 3-day Healthcare Datathon.

※Note: Keynote speeches are open to the public. Registration link will be announced soon.

Keynote speeches registration link: (已報名「2019 TMU-MIT Healthcare Datathon醫療數據松」之參賽者不需再另外報名此系列講座)

Day1: https://www.accupass.com/event/1907030820491391814673

Day2: https://www.accupass.com/event/1907080700164653196780

※Keynote Speech

Topic: Time-Limited Trials for Critically Ill Cancer Patients

Speaker: Peggy Lai

There are many clinical questions that are difficult to answer, because conducting a trial would be unethical while observational studies cannot address unmeasured or unmeasurable confounders. Careful application of innovative approaches to "big data" collected through electronic health records may fill some of these gaps. When a patient with poor long-term prognosis (as in the case of advanced cancer) becomes critically ill, with uncertain short-term prognosis determined more by severity of organ failure than underlying disease, what should a clinician recommend? We applied decision-analytic methods to three large intensive care unit databases to ask the question, "If I have a poor-prognosis cancer and become critically ill, how long of a trial of intensive care would give me the same shot at surviving 30 days as time-unlimited care?" (PMID: 26469222).

※Keynote Speech

Topic: When Machine Learning Meets Healthcare

Speaker: Mengling 'Mornin' Feng

Machine Learning (ML), especially the application of Artificial Intelligence (AI), is no doubt one of the most popular research filed at the moment. Specifically for healthcare, many believe that the current advancement in ML and AL is going to augment and disrupt the current practices so to achieve better and more cost effective care. In my talk, I will share a number of real use cases that my group have been working on in the past years, where ML and AI technologies were applied to address healthcare problems. More importantly, I will also share our realisations on the major challenges and limitations while deploying ML and AI solutions for real clinical applications.

※Keynote Speech

Topic: AI in Healthcare: Enhancing the Efficacy and Efficiency of Interactions to the Rural World

Speaker: Rani Shifron

Global Medicine has turned the World Flat - Technologies are flattening the global provision and extend healthcare access and affordability. Today's Technologies enhance the efficacy and efficiency of interactions among people who are geographically dispersed.

Technologies that approximate Presence by enabling interaction in Mixed Reality between two parties, who are geographically apart, to allow for pin-pointing and finger-pointing on an object, a document or, in the case of Radiology, an X-ray Film for clarity in understanding and accuracy/precision in description and thereby avoiding errors.

Technologies that allow for doing routine checkups at home or in remote locations and those that can help monitor high risk patients in real time.

As an example VR (Virtual Reality) Streaming and MR (Mixed Reality) Streaming with the advent of 5G cellular coverage will be an industry disruptor. The rural world is getting 5G before the western world and that is why this is our chance to allow the rural world to play catch up with the rest of the world.

I will give some examples and discuss what I think the short and longer term future will bring.

※Note: All participants are required to sign NDA (Non-Disclosure Agreement) for database download.

※Note: Database download access will be ready soon and download via GCP (Google Cloud Platform).

(https://mimic.physionet.org/)

MIMIC is an openly available dataset developed by the MIT Lab for Computational Physiology, comprising deidentified health data associated with 40,000 critical care patients. It includes demographics, vital signs, laboratory tests, medications, and more.

(https://mimic.physionet.org/)

The eICU Collaborative Research Database is a large multi-center critical care database made available by Philips Healthcare in partnership with the MIT Laboratory for Computational Physiology. The eICU Collaborative Research Database holds data associated with over 200,000 patient stays, providing a large sample size for research studies.