Objectives
Individualized treatment for bipolar disorder based on neuroimaging treatment targets remains elusive. To address this shortcoming, we developed a linguistic machine learning system based on a cascading Genetic Fuzzy Tree (GFT) design called the LITHium Intelligent Agent (LITHIA). Using multiple objectively defined fMRI and proton MRS (1H-MRS) inputs, we tested whether LITHIA could accurately predict lithium response in participants with first-episode bipolar mania.
Methods
We identified 20 subjects with first-episode bipolar mania who received an adequate trial of lithium over eight weeks and both fMRI and 1H-MRS scans at baseline pre-treatment. We trained LITHIA using 18 1H-MRS and 90 fMRI inputs over four training runs to classify treatment response and predict symptom reductions. Each training run contained a randomly selected 80% of the total sample and was followed by a 20% validation run. Over a different randomly selected distribution of the sample, we then compared LITHIA to eight common classification methods.
Results
LITHIA demonstrated nearly perfect classification accuracy and was able to predict post-treatment symptom reductions at 8 weeks with at least 88% accuracy in training and 80% accuracy in validation. Moreover, LITHIA exceeded the predictive capacity of the eight comparator methods and showed little tendency towards overfitting.
Conclusions
Results provide proof-of-concept that a novel GFT is capable of providing control to a multidimensional bioinformatics problem – namely prediction of lithium response – in a pilot data set. Future work on this, and similar machine learning systems, could help assign psychiatric treatments more efficiently, thereby optimizing outcomes and limiting unnecessary treatment.
This study introduces the technique of Genetic Fuzzy Trees (GFTs) through novel application to an air combat control problem of an autonomous squadron of Unmanned Combat Aerial Vehicles (UCAVs) equipped with next-generation defensive systems. GFTs are a natural evolution to Genetic Fuzzy Systems, in which multiple cascading fuzzy systems are optimized by genetic methods. In this problem a team of UCAV's must traverse through a battle space and counter enemy threats, utilize imperfect systems, cope with uncertainty, and successfully destroy critical targets. Enemy threats take the form of Air Interceptors (AIs), Surface to Air Missile (SAM) sites, and Electronic WARfare (EWAR) stations. Simultaneous training and tuning a multitude of Fuzzy Inference Systems (FISs), with varying degrees of connectivity, is performed through the use of an optimized Genetic Algorithm (GA). The GFT presented in this study, the Learning Enhanced Tactical Handling Algorithm (LETHA), is able to create controllers with the presence of deep learning, resilience to uncertainties, and adaptability to changing scenarios. These resulting deterministic fuzzy controllers are easily understandable by operators, are of very high performance and efficiency, and are consistently capable of completing new and different missions not trained for.
Fuzzy logic is used in a variety of applications because of its universal approximator attribute and nonlinear characteristics. But, it takes a lot of trial and error to come up with a set of membership functions and rule-base that will effectively work for a specific application. This process could be simplified by using a heuristic search algorithm like Genetic Algorithm (GA). In this paper, genetic fuzzy is applied to the task assignment for cooperating Unmanned Aerial Vehicles (UAVs) classified as the Polygon Visiting Multiple Traveling Salesman Problem (PVMTSP). The PVMTSP finds a lot of applications including UAV swarm routing. We propose a method of genetic fuzzy clustering that would be specific to PVMTSP problems and hence more efficient compared to k-means and c-means clustering. We developed two different algorithms using genetic fuzzy. One evaluates the distance covered by each UAV to cluster the search-space and the other uses a cost function that approximates the distance covered thus resulting in a reduced computational time. We compare these two approaches to each other as well as to an already benchmarked fuzzy clustering algorithm which is the current state-of-the-art. We also discuss how well our algorithm scales for increasing number of targets. The results are compared for small and large polygon sizes.
The problem of assigning a group of Unmanned Aerial Vehicles (UAVs) to perform spatially distributed tasks often requires that the tasks will be performed as quickly as possible. This problem can be defined as the Min–Max Multiple Depots Vehicle Routing Problem (MMMDVRP), which is a benchmark combinatorial optimization problem. In this problem, UAVs are assigned to service tasks so that each task is serviced once and the goal is to minimize the longest tour performed by any UAV in its motion from its initial location (depot) to the tasks and back to the depot. This problem arises in many time-critical applications, e.g. mobile targets assigned to UAVs in a military context, wildfire fighting, and disaster relief efforts in civilian applications. In this work, we formulate the problem using Mixed Integer Linear Programming (MILP) and Binary Programming and show the scalability limitation of these formulations. To improve scalability, we propose a hierarchical market-based solution (MBS). Simulation results demonstrate the ability of the MBS to solve large scale problems and obtain better costs compared with other known heuristic solution.
Utilizing a combination of heuristic and stochastic techniques, this work approximates solutions for complex variants of the Traveling Salesman Problem. Within this study, we seek to develop a Pareto front of time optimal solutions for a large scale, multi-objective, multi-agent, multi-depot, traveling salesman problem with turning constraints that emulates a UAV operator monitoring a swarm of autonomous vehicles. This research is a culmination of years of investigation and past publications [1, 2, 3] which sought to solve portions of this problem individually. Solving the entire problem, more precisely named the Multi-Objective Min-Max Multi-Depot Polygon Visiting Dubins Multiple Traveling Salesman Problem, follows similar processes of simpler variants. With a focus on military applications, run-time is very critical to the success of this software. Through the cooperation of these algorithms, accurate solutions for large scale problems can be produced within a quick timeframe.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.