High Tech Consortium of Southern New Mexico Monthly [July] Meeting & Speaker. We have two speakers this month. SPEAKER TOPIC I: A Decision Support System for Integrated Surgical Scheduling. SPEAKER TOPIC II: Rule-Based Classification for Big Data: A Heuristic Approach.

The July meeting and speaker of the High Tech Consortium (HTC) of Southern New Mexico will be on the 24th of July [Wednesday], from 4:30 p.m. to 6:00 p.m. at the Arrowhead Center [located at 3655 Research Drive, Genesis Center, Building C, on the campus of New Mexico State University].

Free to attend.

Our first speaker is Sajia Afrin Ema. She is a PhD candidate in Industrial Engineering Department at New Mexico State University. Her research interest is in developing decision support systems for optimization of healthcare services including surgery and outpatient scheduling. Her expertise includes developing mathematical and simulation models, implementation of heuristic algorithm and statistical inferencing for healthcare systems.

A Decision Support System for Integrated Surgical Scheduling: Surgical planning and scheduling deals with optimal assignment of scarce and expensive hospital resources. The stochastic nature of the problem and involvement of multiple stakeholders with often conflicting priorities makes this task complicated to accomplish. Over the last few decades, several operations research studies have been conducted to facilitate the surgical scheduling process, however, most of the researches considered the deterministic versions of the problem which did not accounted for the dynamicity involved in the surgical scheduling process such as arrival of emergency patient, variability in surgical duration and change of patient priorities over time and so on.  In our research so far, we have developed a mathematical model that has the capability to optimally co-handle both elective and emergency surgeries using a common set of surgical resources including surgeons, operating rooms and Operating Room nurses. The overall objective of the model aims at improving the utilization of surgical resources and providing the patients with timely access to necessary care. The ongoing work involves inclusion of other downstream units to the model for validation and development of an Ant Colony Optimization Algorithm to solve larger version of the problem in efficient manner.

Our second speaker is Sayed Kaes M Hossain. He is a Ph.D. student in the Industrial Engineering Department at New Mexico State University. His area of expertise includes- Combinatorial Optimization, Heuristic Algorithms, and Data Mining. In the recent past, he has developed optimization models for Energy Systems, Assembly Lines and Facility Layout problems.

Rule-Based Classification for Big Data: A Heuristic Approach: In recent years, a handful number of machine learning algorithms have been developed for data mining in large datasets. In many cases, these algorithms yield a very complex classifier in order to provide high accuracy. However, such complex classifiers are sometimes difficult to interpret. Applications those require an understanding of the classifier itself in order to integrate the human knowledge and the knowledge extracted from data, demands a classifier that provides acceptable accuracy, yet easily interpretable. Rule-based classifiers seem to fit in this role. However, exploring such rules from a big dataset can be computationally expensive. Thankfully, we can incorporate our expertise from combinatorial optimization to address this issue. We are developing a modified classification algorithm based on a well-known heuristic algorithm, Ant Colony Optimization algorithm (ACO). The earlier applications of ACO in classification are known as Ant Miner algorithms. We are proposing a dynamic tuning of the balance between exploration and exploitation as the algorithm progresses. A functional module based on this idea is already implemented in our laboratory. In the next phase, we are implementing the competing algorithms to evaluate the relative performance of our algorithm.

HTC membership meetings are open to the public. Anyone interested in growing technology in southern New Mexico is encouraged to attend. There is no charge for attending. For more information please call the HTC President, Dr. Ed Pines, at 575-646-2730, or Terry Jack 720-201-7344.

VISIT our website at: www.HTCNM.com




We have 275 guests and no members online

Latest Post