This document discusses using cohort and trajectory analysis in multi-agent support systems to provide personalized assistance to cancer survivors. It describes how combining several machine learning models can extract a patient's risk level and predict their likelihood of survival. For example, trajectory analysis of tumor stage 2 patients found an 80% chance of surviving over 5 years. Overall, the research aims to develop personalized chatbot and mobile app assistants that integrate a patient's electronic health records and behavior data with hybrid machine learning models to conduct advanced cohort and trajectory analysis for improved clinical decision support and quality of life support.
Energy Resources. ( B. Pharmacy, 1st Year, Sem-II) Natural Resources
Manzo_a2hc_aamas
1. Cohort and Trajectory Analysis
in Multi-Agent Support Systems
for Cancer Survivors
Gaetano Manzo, Davide Calvaresi, Oscar Jimenez del Toro, Jean-Paul Calbimonte,
and Michael Schumacher
University of Applied Sciences Western Switzerland
(HESSO)
A2HC WORKSHOP – AAMAS 2021
2. 281,550 new cases of invasive breast cancer are
estimated in the U.S. in 2021
G. Manzo A2HC 2021 2
all annual cancer
cases
Of patients are
women
Of cases are < 50
years old
50%
Patients must cope with physical and psychological sequelae
12% 99%
9. Conclusion and Future work
G. Manzo A2HC 2021 9
Personalized
Support for
breast cancer
patients
EREBOTS,
Cohort and
trajectory
analysis
Hybrid
models and
risk markers
detection
Inject EHRs
and
Behavioural
data
Advanced
CTA
model
selection
HEMERA app
Editor's Notes
Breast cancer
Physical and metal health
Support system
Chatbots
Survivor models
Challenges in the paper
personalized support and assistance after discharge maylead to a rapid diminution of their physical abilities, cognitive impairment,and reduced quality of life.
CTA for EREBOTS
EXTRACT PATIENTS' TRAJECTORY
EXAPLANATION OF THE FEATURE
ADD AUTOMATED MODEL EVALUATED WITH C-INDEX METRICS
ADD PATIENT BEHAVIOURAL DATE