ISSN 0147-6882, Scientific and Technical Information Processing, 2010, Vol. 37, No. 6, pp. 407–412. © Allerton Press, Inc., 2010.
Original Russian Text © V.M. Khachumov, 2009, published in Iskusstvennyi Intellekt i Prinyatie Reshenii, 2009, No. 3, pp. 25–32.
407
INTRODUCTION
Application of new information technologies in
clinical medicine leads to increased efficiency of treat-
ment when these provide the operative management of
treatment process. This is possible if the process of
treatment is considered as a medical technological
process (MTP) [1]. This approach includes the devel-
opment of methods and analytical software of prece-
dents of therapeutic and diagnostic processes for each
of a group of nosological forms, stored in databases of
hospital information systems, building a generalized
description of clinical processes, and adaptation of
such descriptions based on particular patient features.
Analysis of the obtained information on the number of
MTPs of disease treatment shows that there are sub-
stantial reserves to increase productivity and quality of
care, improve the use of medical equipment, etc. The
more fully an MTP accounts for relationships and
constraints linking the conditions of the process with
its yield characteristics, the higher efficiency it has.
Unfortunately, almost all separately taken MTPs are
based on incomplete knowledge of the relationships,
therefore, they can’t provide optimal treatment of
patients.
One of the most effective forms of MTP implemen-
tation is a therapeutic schedule, which after filling is
soon able to serve patients [2]. Various operations of
MTPs may differ in their time of implementation and
equipment used. The length of the schedule is deter-
mined by its longest operation, so it can be down due
to the irregular loading of the schedule. At the same
time, different MTPs can partially use the same equip-
ment, which creates a problem of optimization of their
joint operation schedule. Since the same disease is
revealed differently in different patients, each of them
can be assigned an individual MTP with a different list
and number of equipment, and sequence of their
application depending on the characteristics of dis-
ease. All these issues lead to the problem of optimiza-
tion of time-ordered MTPs. To substantially increase
the efficiency of an MTP, new methods of medical
process analysis are required, based on the automated
extraction of knowledge from existing MTP databases,
using queuing theory and scheduling theory, and the
application of intelligent technologies for decision
support.
The ultimate goal of this research is to develop a set
of specialized optimization methods of medical and
technological treatment processes. The complex
should support an interactive mode with a physician’s
smart graphical user interface and provide the ability
to actively influence the course of the MTP. In this
paper we consider one of the possible approaches to
the formalization of the description and optimization
of an ophthalmic time-ordered MTP.
1. BASICS OF FORMALIZATION AND
OPTIMIZATION OF TIME-ORDERED MTPS
The problem of technological process design is
polyvariant. In an MTP, the number of possible transi-
tion combinations, operation modes and configura-
tions increases significantly for simple diseases and
even more for complex ones. Because of differences in
the structure, equipment used and modes, MTP
options for one disease have different yield character-
istics: the time of treatment, cost, and coefficient of
equipment loading, etc. The presence of several MTP
variants leads to the problem of choosing the best one,
called the optimal one. To reduce the time required to
solve the problem, selection of rational variants of
MTP design decisions is used. This leads to the prob-
lem of forming the criteria of options selection.
A pipelined MTP in the framework of scheduling
theory is formalized in the form of a system with mul-
tiphase cyclical maintenance. Periodic schedules are
best suited for the planning of multiple execution of
complex operations. In [3–5] an approach was devel-
oped to the design of time-ordered schedules. Let us
consider the main points of this approach based on the
optimization of combining service cycles during mul-
tiple realizations of MTPs.
Intellectual Systems Models of Scheduling Medicinal Processes
V. M. Khachumov
Abstract—A method of formalization and optimization of schedules for medical process is considered.
Examples of optimization of medical schedules of ophthalmic hardware aid are given. We propose artificial
intelligence methods to improve the quality of patient care.
Keywords: healing schedule, medical process, the combination of cycles, optimization, artificial intelligence.
DOI: 10.3103/S0147688210060055