• 沒有找到結果。

Smart Healthcare

N/A
N/A
Protected

Academic year: 2022

Share "Smart Healthcare "

Copied!
79
0
0

加載中.... (立即查看全文)

全文

(1)

智慧醫療

Smart Healthcare

臺北市立聯合醫院仁愛院區家庭醫學科 郭冠良

Kuan-Liang Kuo, MD, PhD

2022-03-30

Ver. 0329

1

(2)

Contents

• 醫療資訊之歷史演進

• 智慧醫療

2

(3)

歷史演進

3

(4)

~2000

• 管理資訊系統 (Management Information System)

– 藥品、衛材、帳務 – ERP

• HIS (Hospital Information System)

– 用途演進?

4

(5)

ERP

BPR ERP EA

CRM

KM EA

EA

EA 供

應 商

經 銷

SCM SCM 商

客戶

EAI EAI

ERP: Enterprise Resource Planning BPR: Business Process Reengineering SCM: Supply Chain Management

EA: Enterprise Application EAI: EA Integration

CRM: Customer Relationship Management

5

(6)

6

RIS

LIS

ERP

PACS Clients

CPOE 狹義的 HIS

廣義的HIS ~2000

住院、急診、門診

CPOE: Computerized Physician Order Entry

(7)

2000~

• HIS

– 整合各部門系統 (廣義的HIS)

– 結合CDSS (Clinical Decision Support System)

7

(8)

Quiz

• 廣義的HIS包含哪些子系統?

8

(9)

2010~

• EMR (Electronic Medical Record)

• Wireless network

• Wearable device

• Smart phone

• IoT (Internet of Things)

9

(10)

EMR

• EMR (Electronic Medical Record)

– 電子簽章 – 無紙化

10

2010~

(11)

Meaningful Use (2010~)

11

2010~

(12)

MIPS (2015~)

(Merit-based Incentive Payment System)

• MIPS harmonizes existing CMS quality programs

– Meaningful Use (MU)

– Physician Quality Reporting System (PQRS) – Value-Based Payment Modifiers (VBPM)

• MIPS consolidates multiple, quality programs into a single program to improve quality care

12

2010~

https://www.healthit.gov/topic/meaningful-use-and-macra/meaningful-use

Merit:價值

(13)

MIPS (2015~)

(Merit based Incentive Payment System)

• The Four Pillars of MIPS Reporting

• Using Certified EHR Technology

13

2010~

https://www.healthit.gov/topic/federal-incentive-programs/MACRA/merit-based-incentive-payment-system

(14)

RIS LIS PACS

Gateway Server:

•XML

•Signature ERP

CPOE

EEC EMR

Data source

電子病歷 2010~

14

衛福部提供軟體:

EEC Gateway

EEC Lite Gateway

(15)

15

(16)

Quiz

• 電子病歷與病歷電子化的不同?

16

(17)

Wireless network 2010~

17

功耗、傳輸速度、傳輸距離

(18)

Wearable Device 2010~

http://www.c-sharpcorner.com/UploadFile/f88748/internet-of-thingsiot-part-4-network-protocols-and-arc/

18

(19)

IoT (Internet of Things)

• IoT Layers

2010~

https://www.c-sharpcorner.com/UploadFile/f88748/internet-of-things-part-2/

19

(20)

IoT (Internet of Things)

• Ontology Hierarchy of IoT

2010~

http://www.c-sharpcorner.com/UploadFile/f88748/internet-of-things-iot-part-3/

20

(21)

IoT (Internet of Things) 2010~

http://dx.doi.org/10.4236/ait.2016.62002

21

(22)

Quiz

• IoT的網路需求

– 網路傳輸頻寬 – 網路傳輸距離 – 功耗

22

(23)

2015~

• Smart Healthcare

– ICT integration with healthcare

• Smart Hospital

– Smart healthcare

– Smart environment / building

• IoT

• AIoT

ICT: Information communication technology

AI: Artificial Intelligence 23

(24)

2015~

• AI in medicine

– ML – DL – …

• Blockchain

– Cryptography

AI: Artificial Intelligence ML: Machine Learning DL: Deep Learning

24

(25)

智慧醫療

Smart Healthcare

25

(26)

Get Smart for Healthcare

• Terms

– Connected health – Digital health

– eHealth – mHealth – iHealth

– Smart healthcare – Digital medicine

26

(27)

• 2018年我國65歲以上人口達14%,依據 WHO定義,已邁入「高齡社會」

27

https://www.ndc.gov.tw/Content_List.aspx?n=695E69E28C6AC7F3 (2019-02-10擷取自國家發展委員會)

Motivation

Smart

Healthcare

(28)

• 出生數下降

– 1968: 40萬 – 2018: 18萬

• 照顧負擔快速升高

28

https://pop-proj.ndc.gov.tw/download.aspx?uid=70&pid=70 (2019-02-10擷取自國家發展委員會)

Motivation

Smart

Healthcare

(29)

Motivation

• Healthcare or Sickcare?

• Health Cost

– 健保預算 ?/年

Smart

Healthcare

• Unfortunately, our healthcare system is a system for the sick

29

Marvasti FF, Stafford RS. From

“Sick Care” to Health Care:

Reengineering Prevention into the U.S. System. The New England journal of medicine.

2012;367(10):889-891.

doi:10.1056/NEJMp1206230.

(30)

Motivation

Smart

Healthcare

Increasing complexity

of

healthcare system

Big data

Medical science becomes

Data-intensive science

Smart healthcare

30

(31)

Motivation

Smart

Healthcare

• P4-medicine

(Lee Hood, MD, PhD)

31 Big data tracking multiple

dimensions of individual health

Ongoing systems analytics

(Systems biology)

Personalized, actionable health information

Data Method P4-medicine

(32)

Motivation

Smart

Healthcare

• P4-medicine

32 Big data tracking multiple

dimensions of individual health

Ongoing systems analytics

(Systems biology)

Personalized, actionable health information

Data Method P4-medicine

(33)

Motivation

Smart

Healthcare

• Systems biology 1 2

Biology Information technology

Molecular, cellular, and phenotypic data Information-based

Systems approaches

Holistic understanding of complex biological

systems 3 Systems biology

Comprehensive, dynamic analyses of the components of a system

Disease and wellness as a continuum of network states unique in time and space to each individual human being

33 Big data tracking multiple

dimensions of individual health

Ongoing systems analytics

(Systems biology)

Personalized, actionable health information

Data Method P4-medicine

(34)

Motivation

Smart

Healthcare

• P4-medicine願景

– Personalized actionable information

• De-identified

• Aggregated

– Physicians

• Make more cost-effective diagnosis and treatments

– Patients (sick people) and consumers (healthy people)

• Live healthier, more productive lives

34

(35)

Motivation

Smart

Healthcare

35

(36)

Get Smart for Healthcare

• Information at point-of-care (POC)

• Guidelines embedded workflow

• Ubiquitous computing

– Network, sensor, mobile device, …

• Measurable healthcare

– Data, data, data

How to?

36

(37)

• Seamless sensors and interactive technologies

• Offer personalized and context- adapted medial support

– Assistance to carry out everyday activities – Monitoring personal health conditions

– Enhancing patient safety

– Getting access to social medical and emergency systems on demand

Ubiquitous Computing

Smart

Healthcare

Comfort, Autonomy,

Detection, Prediction, Prevention

Context  data, data, data

37

(38)

Ubiquitous Computing

Smart

Healthcare

http://lmeasy.blogspot.tw/2015/11/ubiquitous-computing-ubicomp.html

38

(39)

Ubiquitous Computing

Smart

Healthcare

http://blogs.parc.com/2010/03/defining-ubiquitous-computing-vs-augmented-reality/

39

• Ubiquity

• Natural interaction

• Proactivity

(40)

Human-in-the-loop

Smart

Healthcare

40

• Cognitive science(人) + Computer science(機)

• Role of humans

– Human as operator – Human as data source

• Cyber-physical system(CPS)

– Interact seamlessly

• 人、機、物的無縫融合計算

• 實體裝置的互動網路

– Adaptive behavior of all

– Intelligence of all

(41)

Human-in-the-loop Smart Healthcare

Generate rules

Holzinger, Andreas, Carsten Röcker, and Martina Ziefle. "From smart health to smart hospitals." Smart health. Springer International Publishing, 2015. 1-20.

41

(42)

Cyber-physical system (CPS)

Smart

Healthcare

42

(43)

• Mobility

• Invisibility

• Natural communication

Technological Challenges

Smart

Healthcare

43

(44)

• Hardware

• Network

• Software

• Data

• Service

Components of Smart Healthcare

Smart

Healthcare

44

(45)

• Device (or sensor)

– Connectivity – Data capture

• Clinical data

• Operational data

– Mobility (miniature, light-weight) – Low power consumption

– Low cost

– FDA approvement (medical device)

Hardware

Smart

Healthcare

45

(46)

• Smart healthcare is built on smart networks

– Ubiquitous

– Combination of multiple protocols

– Q&A: Bandwidth issue, security issue

Network

Smart

Healthcare

46

See next slide

(47)

• Scenario example:

– Range vs energy consumption

•BLE

•Xenon

•Wi-Fi

•LAN

Internet

•GPRS

•3G/4G

BLE: Bluetooth Low Energy LAN: Local Area Network

Network

Smart

Healthcare

47

Combination of multiple protocols

(48)

Network

• Range, Power, and Data rates

– 4G/Wi-Fi/BLE

48

http://embedur.com/blogsandnews/Choosing-the-right-wireless-tech-for-IoT.html BLE: Bluetooth Low Energy

(49)

• 智慧醫療最重要,最核心部分

• Software

– 1.應用程式

– 2.背景服務程式 (伺服器) – 3.批次處理程式

– 4.API程式

Software

Smart

Healthcare

API: Application Programming Interface 49

(50)

• 應用程式

– User Interface

• Data

• Control

• POC

– Logic

• Workflow

• CDSS

– Database

Software

Smart

Healthcare

MVC Model

50

1.應用程式

(51)

51

(52)

• Types of application framework

– Standalone – Client-server

• Thin-client

• Fat-client

Software

Smart

Healthcare

52

1.應用程式

(53)

• Thin-client

– Server-based computing

• Web-based computing

• Cloud-computing

– Virtual desktop

Software

Smart

Healthcare

53

Client-server

• Thin-client

• Fat-client

http://www.atlanticcs.net/cloudcomputing.html

http://www.sfsolutions.co.th/solutions/virtual-desktop-vdi

1.應用程式

(54)

• Fat-client

– PC-based

Software

Smart

Healthcare

54

Client-server

• Thin-client

• Fat-client

1.應用程式

(55)

• Thin-client v.s. Fat-client

Software

Smart

Healthcare

55

Client-server

• Thin-client

• Fat-client

http://paransilverlight.tistory.com/258

Pros and Cons?

1.應用程式

(56)

• 背景服務程式

– 作業系統 (OS) – 伺服器 (server)

– 排程啟動程式 (cron)

Software

Smart

Healthcare

56 https://msdnshared.blob.core.windows.net/media/2017/01/servers.jpg

2.背景服務程式

(57)

• 批次處理程式

– 資料處理

– 與排程啟動程式結合

Software

Smart

Healthcare

57

https://image.slidesharecdn.com/april131220dataartisanscelebi- 160425200014/95/unified-stream-and-batch-processing-with-apache- flink-7-638.jpg?cb=1461614428

3.批次處理程式

(58)

• API程式

– 分散式計算的解決方案

• 平行計算 /計算量分工

– SaaS (Software as a service)

• Web service

• 例:

– AWS (Amazon web services)

– Google cloud

Software

Smart

Healthcare

58

4.API程式

https://image.slidesharecdn.com/april131220dataartisanscelebi-

160425200014/95/unified-stream-and-batch-processing-with-apache-flink-7- 638.jpg?cb=1461614428

https://coderwall.com/p/klzs5a/hello-distributed-computing

(59)

• Source

– HIS

– Medical device / IoT

• Type

– Clinical data

– Operational data

• Applications

– Alarm management – Patient surveillance

– Asset management (Extended ERP)

• Purpose

– Improve clinical outcome

– Enhance workflow and operational efficiency

Data

Smart

Healthcare

59

Medical Science vs

Data Science

(60)

• AI in medicine

– Rule-based

• Expert system

– Statistical method

• Bayesian network

– Neural network (1943~)

• Machine learning

• Deep learning

• Clinical Decision Support System

– Alert

– Reminder – Workflow

Data

Smart

Healthcare

60

Data Science

(61)

Data

Smart

Healthcare

61

Public Key Encryption and Digital Signature: How do they work?

2004 by CGI Group Inc.

(62)

• Blockchain

– Chain and data

Data

Smart

Healthcare

62

(63)

Data

Smart

Healthcare

63

• Blockchain

– Security

(64)

• Blockchain Technology

– Distributed network – Shared ledger

– Digital transaction

• Advantages of Healthcare blockchain

– Open standard

– For researchers, healthcare providers, care givers, and individuals

– Fault tolerance and disaster recovery – Keep lifetime data

– Change clinical practice

• From guideline to personalized treatment path

Data

Smart

Healthcare

64

•Scalability

•Access security and data privacy

(65)

• Based on

– Humanity

– Person-centered

• HCAHPS (Hospital Consumer Assessment of Healthcare Providers and Systems)

Service

Smart

Healthcare

65

Comfort, Autonomy, Detection, Prediction, Prevention

HCAHPS: http://www.hcahpsonline.org

(66)

• For

– Point-of-care

• Patient

• Clinician / Healthcare provider

– Institute

– Public health – Policy

Service

Smart

Healthcare

66

(67)

• Scenarios

– Medical education

• Simulation

– Patient care

• CDSS

• Workflow

• Outcome

Service

Smart

Healthcare

67

(68)

• Scenarios

– Research

• Clinical data

• Operational data

– Hospital operation

• Quality and efficiency

Service

Smart

Healthcare

68

(69)

• Learning Health System (LHS)

– Rapid Increasing Complexity

– Achieving Greater Value in Health Care

– Capturing Opportunities

from Technology, Industry, and Policy

Service

Smart

Healthcare

69

(70)

• Learning Health System (LHS) (李友專)

– Data to Knowledge

– Knowledge to Practice – Practice to Consumer – Consumer to Data

Service

Smart

Healthcare

70

(71)

期程 易感受期 臨床前期 臨床期 殘障期 生命末期與臨終期

初段 次段 三段 四段

級別 第一級 第二級 第三級 第四級 第五級 第六級 第七級

健康促進

特殊防護

早期診斷 早期治療

限制殘障

居家照護

緩和醫療

安寧照護

介入方案

健康促進活動 勞工健檢 健康檢查 慢性病個案管

居家復健 居家安寧 在宅臨終 營養衛教 特殊族群健檢 四癌篩檢 失智照護 居家照護 圓夢計畫 悲傷輔導 健康識能 居家防跌 失智篩檢防治 用藥整合 在宅服務

預防注射 老人憂鬱篩檢 住院治療

輔具運用 出院準備

推展政策

家庭責任醫師 家庭責任醫師 家庭責任醫師 家庭責任醫師 居家醫療整合 (S0, S1, S2)

居家醫療整合 (S3)

居家醫療整合 (S3) ACP/AD ACP/AD

SDM ACP/AD 開放醫院

SDM ACP/AD

PAC

SDM ACP/AD

SDM DNR/AD

SDM DNR

四段七級的預防醫學

Smart

Healthcare

(72)

期程 易感受期 臨床前期 臨床期 殘障期 生命末期與臨終期

初段 次段 三段 四段

級別 第一級 第二級 第三級 第四級 第五級 第六級 第七級

健康促進

特殊防護

早期診斷 早期治療

限制殘障

居家照護

緩和醫療

安寧照護

介入方案

健康促進活動 勞工健檢 健康檢查 慢性病個案管

居家復健 居家安寧 在宅臨終 營養衛教 特殊族群健檢 四癌篩檢 失智照護 居家照護 圓夢計畫 悲傷輔導 健康識能 居家防跌 失智篩檢防治 用藥整合 在宅服務

預防注射 老人憂鬱篩檢 住院治療

輔具運用 出院準備

推展政策

家庭責任醫師 家庭責任醫師 家庭責任醫師 家庭責任醫師 居家醫療整合 (S0, S1, S2)

居家醫療整合 (S3)

居家醫療整合 (S3) ACP/AD ACP/AD

SDM ACP/AD 開放醫院

SDM ACP/AD

PAC

SDM ACP/AD

SDM DNR/AD

SDM DNR

四段七級的預防醫學

Smart

Healthcare

減重、營養、衛教(EQ App鼓勵互動) 戒菸(IoT,EQ App鼓勵互動)

預注提醒

跌倒(平衡、環境)監測IoT 特殊工作環境IoT

智慧輔具

(73)

期程 易感受期 臨床前期 臨床期 殘障期 生命末期與臨終期

初段 次段 三段 四段

級別 第一級 第二級 第三級 第四級 第五級 第六級 第七級

健康促進

特殊防護

早期診斷 早期治療

限制殘障

居家照護

緩和醫療

安寧照護

介入方案

健康促進活動 勞工健檢 健康檢查 慢性病個案管

居家復健 居家安寧 在宅臨終 營養衛教 特殊族群健檢 四癌篩檢 失智照護 居家照護 圓夢計畫 悲傷輔導 健康識能 居家防跌 失智篩檢防治 用藥整合 在宅服務

預防注射 老人憂鬱篩檢 住院治療

輔具運用 出院準備

推展政策

家庭責任醫師 家庭責任醫師 家庭責任醫師 家庭責任醫師 居家醫療整合 (S0, S1, S2)

居家醫療整合 (S3)

居家醫療整合 (S3) ACP/AD ACP/AD

SDM ACP/AD 開放醫院

SDM ACP/AD

PAC

SDM ACP/AD

SDM DNR/AD

SDM DNR

四段七級的預防醫學

Smart

Healthcare

健檢篩檢提醒

失智、憂鬱監測IoT、EQ App 互動

慢性病監測IoT 用藥監測IoT

智慧輔具

(74)

Future

• Smart healthcare perspective in hospital information management

– BPR(Business Process Re-engineering)

Q & A

Smart

Healthcare

74

(75)

趨勢

75

(76)

趨勢

• 2017台大

– 生物資訊學國際研究生博士學位學程 – 資料科學碩博士學位學程

• 2018成大

– 數據科學研究所

76

(77)

課程資料

• http://www.csie.ntu.edu.tw/~d93009/scholar.php

77

(78)

興趣 (2) –

Rabees Band

78

(79)

Thanks for Your Attention

79

參考文獻

相關文件

Finally, we train the SLU model by learning latent feature vectors for utterances and slot candidates through MF techniques. Combining with a knowledge graph propagation model based

A spoken language understanding (SLU) component requires the domain ontology to decode utterances into semantic forms, which contain core content (a set of slots and slot-fillers)

“Find sufficiently accurate starting approximate solution by using Steepest Descent method” + ”Compute convergent solution by using Newton-based methods”. The method of

Curriculum planning - conduct holistic curriculum review and planning across year levels to ensure progressive development of students’ speaking skills in content, organisation

化成 reduced echelon form 後由於每一個 row 除了該 row 的 pivot 外, 只剩 free variables (其他的 pivot variable 所在的 entry 皆為 0), 所以可以很快地看出解的形式..

• Competing physics on multiple length & time

List up all different types of high-sym k (points, lines, planes) 2...

SS Curriculum and Assessment for the Ethics and Religious Studies Curriculum Ethics and Religious Studies Curriculum Enhancement of Assessment Literacy:?. Designing and Marking