Patient Journey Optimization using a Multiagent approach Choi

Patient Journey Optimization using a Multi-agent approach Choi Chung Ho 1

Agenda l l l Introduction Problem formulation Scheduling framework Agent coordination Experiments Conclusion 2

Introduction 3

Our goal l To improve patient journey by reducing undesired waiting time for patients 4

How to achieve our goal? l To schedule patients in such a way that medical resources could be utilized in a more efficient manner 5

Why using a multi-agent approach? l Hospitals are found to have a decentralized structure l A multi-agent approach is proposed as it favors the coordination between geographically distributed entities 6

Related works of using a multi-agent approach for patient scheduling l T. O. Paulussen, I. S. Dept, K. S. Decker, A. Heinzl, and N. R. Jennings. Distributed patient scheduling in hospitals. In Coordination and Agent Technology in Value Networks. GITO, pages 1224– 1232. Morgan Kaufmann, 2003. The use of health state as an utility function has been challenged l I. Vermeulen, S. Bohte, K. Somefun, and H. La Poutre. Improving patient activity schedules by multi-agent pareto appointment exchanging. In CEC-EEE ’ 06: Proceedings of the The 8 th IEEE International Conference on E-Commerce Technology and The 3 rd IEEE International Conference on Enterprise Computing, ECommerce, and E-Services, page 9, Washington, DC, USA, 2006. IEEE Computer Society. Temporal constraints between treatment operations are not considered 7

Problem formulation 8

Seven cancer centers in Hong Kong C = {HKE, HKW, KC, KE, KW, NTE, NTW} 9

Treatment operations and medical resources Treatment plan Treatment operations ( ) { Radiotherapy planning, Radiotherapy, Surgery, Chemotherapy } Medical resources (A) { Radiotherapy planning unit, Radiotherapy unit, Operation unit, Chemotherapy unit } 10

Patient journey l We define Patient journey as: Duration from the date of admission to the date of the last treatment operation completed 11

Scheduling framework 12

Two types of agents l l Patient agent Resource agent 13

Patient agent l A patient agent (Pi) is used to represent one cancer patient l Each Pi stores the corresponding patient’s treatment plan Treatment plan 14

Resource agent l A resource agent is used to represent one specific medical unit, denoted as Rab a A, b C Center (HKE) Center (KC) Center (KW) Center (NTW) Radiotherapy planning unit Radiotherapy unit Center (HKW) Center (KE) Center (NTE) Operation unit Chemotherapy unit 15

Scheduling algorithm Pareto improvement 16

Agent coordination 17

Coordination framework 18

Coordination framework (cont. ) l For each request, it includes: l 1) Earliest possible start date (EPS) It is the earliest date on which a treatment operation could start l 2) Latest possible start date (LPS) It is the latest date on which a treatment operation should start such that the treatment operation could be performed earlier 19

Earliest possible start date (EPS) (j – 1) th treatment operation 20

Latest possible start date (LPS) (j – 1) th treatment operation j th treatment operation 1 day 21

Coordination framework (cont. ) 22

Coordination framework (cont. ) l In order to compute the bid value, three binary variables were defined: l 1) Last 2) Noti 3) Temp l l 23

Coordination framework (cont. ) l l Last is a binary variable that specifies whether the involving treatment operation is the last one in PG’s treatment plan; Last = 0 if it is not the last one; otherwise 1 th treatment operation 2 nd treatment operation 3 rd treatment operation 24

Coordination framework (cont. ) l l Noti is a binary variable that specifies whethere is a week’s time of notification for the target patient agent regarding the exchange; Noti = 0 if there is a week’s time of notification; otherwise 25

Coordination framework (cont. ) l l Temp is a binary variable that specifies whether the temporal constraints between treatment operations are violated for the target patient agent after the proposed exchange; Temp = 0 if no violation; otherwise 26

Coordination framework (cont. ) l For each target patient agent PG: 27

Coordination framework (cont. ) Coordination process for eliminating unnecessary exchanges 28

Unnecessary exchanges 29

Experiments 30

Data set l 5819 cancer patients in Hong Kong, with an admission period of 6 months (1/7/2007 – 31/12/2007) l The average length of patient journey is 90. 7 days before applying our framework 31

Experiments (cont. ) l Group A: The scheduled treatment plans in the dataset are used for the initial assignment l Group B: Only the statistics of the scheduled treatment plans and the capacities of medical units are used for the initial assignment 32

Experiment settings l Setting 1) All patient agents are willing to exchange their timeslots with others whenever there is a Pareto improvement l Setting 2) Only 20% of the patients of each center are allowed to exchange their timeslots l Setting 3) Patients are only be swapped to a nearby cancer center l Setting 4) Timeslots released by deceased patients are allocated to those who have the longest patient journey 33

Experimental results Group A Group B 34

Experimental results (cont. ) Group B 35

Conclusion 36

Conclusion and future works l A multi-agent framework has been proposed for patient scheduling l In this framework, while no single patient will get a lengthened patient journey, all the temporal constraints between treatment operations would not be violated 37

Conclusion and future works (cont. ) l Experiments show that the average length of patient journey could be reduced by about a week’s time by using the proposed framework l In the future, we are going to see how the bids submitted by the target patient agent could be defined in a more sophisticated way such that the overall patient journey could be shortened in greater extent 38

The end 39