Browsing by Subject "neural networks"

Now showing 1 - 8 of 8
  • Thumbnail Image
    Item
    An Adversorial Approach to Enable Re-Use of Machine Learning Models and Collaborative Research Efforts Using Synthetic Unstructured Free-Text Medical Data
    (IOS, 2019) Kasthurirathne, Suranga N.; Dexter, Gregory; Grannis, Shaun J.; Epidemiology, School of Public Health
    We leverage Generative Adversarial Networks (GAN) to produce synthetic free-text medical data with low re-identification risk, and apply these to replicate machine learning solutions. We trained GAN models to generate free-text cancer pathology reports. Decision models were trained using synthetic datasets reported performance metrics that were statistically similar to models trained using original test data. Our results further the use of GANs to generate synthetic data for collaborative research and re-use of machine learning models.
  • Thumbnail Image
    Item
    Fuzzy Controller Algorithm for Automated HVAC Control
    (IAARC, 2020-10) Chae, Myungjin; Kang, Kyubyung; Koo, Dan D.; Oh, Sukjoon; Chun, Jae Youl; Engineering Technology, School of Engineering and Technology
    This research presents the design framework of the artificial intelligent algorithm for an automated building management system. The AI system uses wireless sensor data or IoT (Internet of Things) and user's feedback together. The wireless sensors collect data such as temperature (indoor and outdoor), humidity, light, user occupancy of the facility, and Volatile Organic Compounds (VOC) which is known as the source of the Sick Building Syndrome (SBS) or New Building Syndrome because VOC are often found in new buildings or old buildings with new interior improvement and they can be controlled and reduced by appropriate ventilation efforts. The collected data using wireless sensors are post-processed to be used in the neural network, which is trained in accordance with the collected data pattern. When the users of the facility have the control of the building's ventilation system and the AI system is fully trained using the user input, it will mimic the user's pattern and control the building system automatically just as the user wants. In this research, data were collected from 4 different buildings: university library, university cafeteria, a local coffee shop, and a residential house. Fuzzy logic controller is also developed for better performance of the HVAC. Indoor air quality, temperature (indoor and outdoor), HVAC fan speed and heater power are used for fuzzified output. As a result, the framework and simulation model for the energy efficient AI controller has been developed using fuzzy logic controller and the neural network-based energy usage prediction model.
  • Thumbnail Image
    Item
    Global Translation of Machine Learning Models to Interpretable Models
    (2021-12) Almerri, Mohammad; Ben Miled, Zina; Christopher, Lauren; Salama, Paul
    The widespread and growing usage of machine learning models, especially in highly critical areas such as law, predicate the need for interpretable models. Models that cannot be audited are vulnerable to inheriting biases from the dataset. Even locally interpretable models are vulnerable to adversarial attack. To address this issue a new methodology is proposed to translate any existing machine learning model into a globally interpretable one. This methodology, MTRE-PAN, is designed as a hybrid SVM-decision tree model and leverages the interpretability of linear hyperplanes. MTRE-PAN uses this hybrid model to create polygons that act as intermediates for the decision boundary. MTRE-PAN is compared to a previously proposed model, TRE-PAN, on three non-synthetic datasets: Abalone, Census and Diabetes data. TRE-PAN translates a machine learning model to a 2-3 decision tree in order to provide global interpretability for the target model. The datasets are each used to train a Neural Network that represents the non-interpretable model. For all target models, the results show that MTRE-PAN generates interpretable decision trees that have a lower number of leaves and higher parity compared to TRE-PAN.
  • Thumbnail Image
    Item
    Low Cost Gunshot Detection using Deep Learning on the Raspberry Pi
    (IEEE, 2019-12) Morehead, Alex; Ogden, Lauren; Magee, Gabe; Hosler, Ryan; White, Bruce; Mohler, George; Computer and Information Science, School of Science
    Many cities using gunshot detection technology depend on expensive systems that ultimately rely on humans differentiating between gunshots and non-gunshots, such as ShotSpotter. Thus, a scalable gunshot detection system that is low in cost and high in accuracy would be advantageous for a variety of cities across the globe, in that it would favorably promote the delegation of tasks typically worked by humans to machines. A repository of audio data was created from sound clips collected from online audio databases as well as from clips recorded using a USB microphone in residential areas and at a gun range. One-dimensional as well as two-dimensional convolutional neural networks were then trained on this sound data, and spectrograms created from this sound data, to recognize gunshots. These models were deployed to a Raspberry Pi 3 Model B+ with a short message service modem and a USB microphone attached, using a software pipeline to continuously analyze discrete two-second chunks of audio and alert a set of phone numbers if a gunshot is detected in that chunk. Testing found that a majority-rules ensemble of our one-dimensional and two-dimensional models fared best, with an accuracy above 99% on validation data as well as when distinguishing gunshots from fireworks. Besides increasing the safety standards for a city's residents, the findings generated by this research project expand the current state of knowledge regarding sound-based applications of convolutional neural networks.
  • Thumbnail Image
    Item
    Natural language processing of MIMIC-III clinical notes for identifying diagnosis and procedures with neural networks
    (arXiv, 2019) Nuthakki, Siddhartha; Neela, Sunil; Gichoya, Judy W.; Purkayastha, Saptarshi; BioHealth Informatics, School of Informatics and Computing
    Coding diagnosis and procedures in medical records is a crucial process in the healthcare industry, which includes the creation of accurate billings, receiving reimbursements from payers, and creating standardized patient care records. In the United States, Billing and Insurance related activities cost around $471 billion in 2012 which constitutes about 25% of all the U.S hospital spending. In this paper, we report the performance of a natural language processing model that can map clinical notes to medical codes, and predict final diagnosis from unstructured entries of history of present illness, symptoms at the time of admission, etc. Previous studies have demonstrated that deep learning models perform better at such mapping when compared to conventional machine learning models. Therefore, we employed state-of-the-art deep learning method, ULMFiT on the largest emergency department clinical notes dataset MIMIC III which has 1.2M clinical notes to select for the top-10 and top-50 diagnosis and procedure codes. Our models were able to predict the top-10 diagnoses and procedures with 80.3% and 80.5% accuracy, whereas the top-50 ICD-9 codes of diagnosis and procedures are predicted with 70.7% and 63.9% accuracy. Prediction of diagnosis and procedures from unstructured clinical notes benefit human coders to save time, eliminate errors and minimize costs. With promising scores from our present model, the next step would be to deploy this on a small-scale real-world scenario and compare it with human coders as the gold standard. We believe that further research of this approach can create highly accurate predictions that can ease the workflow in a clinical setting.
  • Thumbnail Image
    Item
    Real-Time Embedded Implementation of Improved Object Detector for Resource-Constrained Devices
    (MDPI, 2022-04-13) Ravi, Niranjan; El-Sharkawy, Mohamed; Electrical and Computer Engineering, School of Engineering and Technology
    Artificial intelligence (A.I.) has revolutionised a wide range of human activities, including the accelerated development of autonomous vehicles. Self-navigating delivery robots are recent trends in A.I. applications such as multitarget object detection, image classification, and segmentation to tackle sociotechnical challenges, including the development of autonomous driving vehicles, surveillance systems, intelligent transportation, and smart traffic monitoring systems. In recent years, object detection and its deployment on embedded edge devices have seen a rise in interest compared to other perception tasks. Embedded edge devices have limited computing power, which impedes the deployment of efficient detection algorithms in resource-constrained environments. To improve on-board computational latency, edge devices often sacrifice performance, creating the need for highly efficient A.I. models. This research examines existing loss metrics and their weaknesses, and proposes an improved loss metric that can address the bounding box regression problem. Enhanced metrics were implemented in an ultraefficient YOLOv5 network and tested on the targeted datasets. The latest version of the PyTorch framework was incorporated in model development. The model was further deployed using the ROS 2 framework running on NVIDIA Jetson Xavier NX, an embedded development platform, to conduct the experiment in real time.
  • Thumbnail Image
    Item
    Real-Time Vehicle Detection from Short-range Aerial Image with Compressed MobileNet
    (IEEE, 2019-05) He, Yuhang; Pan, Ziyu; Li, Lingxi; Shan, Yunxiao; Cao, Dongpu; Chen, Long; Electrical and Computer Engineering, School of Engineering and Technology
    Vehicle detection from short-range aerial image faces challenges including vehicle blocking, irrelevant object interference, motion blurring, color variation etc., leading to the difficulty to achieve high detection accuracy and real-time detection speed. In this paper, benefiting from the recent development in MobileNet family network engineering, we propose a compressed MobileNet which is not only internally resistant to the above listed challenges but also gains the best detection accuracy/speed tradeoff when comparing with the original MobileNet. In a nutshell, we reduce the bottleneck architecture number during the feature map downsampling stage but add more bottlenecks during the feature map plateau stage, neither extra FLOPs nor parameters are thus involved but reduced inference time and better accuracy are expected. We conduct experiment on our collected 5-k short-range aerial images, containing six vehicle categories: truck, car, bus, bicycle, motorcycle, crowded bicycles and crowded motorcycles. Our proposed compressed MobileNet achieves 110 FPS (GPU), 31 FPS (CPU) and 15 FPS (mobile phone), 1.2 times faster and 2% more accurate (mAP) than the original MobileNet.
  • Thumbnail Image
    Item
    Route-Sensitive Fuel Consumption Models for Heavy-Duty Vehicles
    (SAE, 2020) Schoen, Alexander; Byerly, Andy; dos Santos, Euzell Cipriano, Jr.; Mechanical and Energy Engineering, School of Engineering and Technology
    This article investigates the ability of data-driven models to estimate instantaneous fuel consumption over 1 km road segments from different routes for different heavy-duty vehicles from the same fleet. Models are created using three different techniques: parametric, linear regression, and artificial neural networks. The proposed models use features derived from vehicle speed, mass, and road grade, which can be easily obtained from telematics devices, in addition to power take-off (PTO) active time, which is needed to capture the power requested by accessories in several heavy-duty vehicles. The robustness of these models with respect to the training data selection is improved by using k-fold cross-validation. Moreover, the inherent underestimation or overestimation bias of the model is calculated and used to offset the fuel consumption estimates for new routes. The study shows that the target application dictates the choice of model features. In fact, the results indicate that depending on the vocation the linear regression and neural network models, which use the same input features, are able to adequately differentiate between the fuel consumption of two vehicles from the same fleet as well as between the fuel consumption of a single vehicle over two different routes. However, the parametric model, which does not utilize PTO active time, is unable to differentiate between two vehicles from the same fleet. This latter model is more suitable for comparing fuel consumption across different fleets of vehicles. In summary, vocation-specific models should be used to optimize fuel consumption for a given fleet of vehicles, whereas general models can only provide insight into aggregated fuel consumption for entire fleets. Moreover, both the accuracy and the precision of the models as measured by their confidence interval should be taken into consideration when comparing fuel consumption estimates for two vehicles from the same fleet or the fuel consumption estimates of an individual vehicle for two different routes. This study shows that the artificial neural network models have narrow 95% confidence intervals and are therefore more precise than the equivalent linear regression models.