Life Sciences Commons^™

Report Of Clinical Bone Age Assessment Using Deep Learning For An Asian Population In Taiwan, Chi Fung Cheng, Eddie Tzung-Chi Huang, Jung-Tsung Kuo, Ken Ying-Kai Liao, Fuu‑Jen Tsai

BioMedicine

Introduction

A deep learning-based automatic bone age identification system (ABAIs) was introduced in medical imaging. This ABAIs enhanced accurate, consistent, and timely clinical diagnostics and enlightened research fields of deep learning and artificial intelligence (AI) in medical imaging.

Aim

The goal of this study was to use the Deep Neural Network (DNN) model to assess bone age in months based on a database of pediatric left-hand radiographs.

Methods

The Inception Resnet V2 model with a Global Average Pooling layer to connect to a single fully connected layer with one neuron using the Rectified Linear Unit (ReLU) activation function consisted of …

Artificial Image Objects For Classification Of Schizophrenia With Gwas-Selected Snvs And Convolutional Neural Network, Xiangning Chen, Daniel G. Chen, Zhongming Zhao, Justin Zhan, Changrong Ji, Jingchun Chen

School of Medicine Faculty Publications

In this article, we propose a new approach to analyze large genomics data. We considered individual genetic variants as pixels in an image and transformed a collection of variants into an artificial image object (AIO), which could be classified as a regular image by CNN algorithms. Using schizophrenia as a case study, we demonstrate the principles and their applications with 3 datasets. With 4,096 SNVs, the CNN models achieved an accuracy of 0.678 ± 0.007 and an AUC of 0.738 ± 0.008 for the diagnosis phenotype. With 44,100 SNVs, the models achieved class-specific accuracies of 0.806 ± 0.032 and 0.820 …

Graphical Models In Reconstructability Analysis And Bayesian Networks, Marcus Harris, Martin Zwick

Systems Science Faculty Publications and Presentations

Reconstructability Analysis (RA) and Bayesian Networks (BN) are both probabilistic graphical modeling methodologies used in machine learning and artificial intelligence. There are RA models that are statistically equivalent to BN models and there are also models unique to RA and models unique to BN. The primary goal of this paper is to unify these two methodologies via a lattice of structures that offers an expanded set of models to represent complex systems more accurately or more simply. The conceptualization of this lattice also offers a framework for additional innovations beyond what is presented here. Specifically, this paper integrates RA and …

An Exploration Of Reinforcement Learning In Complex Environments, Karsten Boettcher

Student Research Symposium

This research discusses the performance of artificial intelligence in completing tasks within a complex, simulated environment. The research provides an introduction into reinforcement learning and how these methods were used to train simulated vehicular agents in the video game Rocket League to perform tasks relating to common in-game objectives. Results are evaluated to consider how the agents performed and what can be learned from the results. The conclusions drawn show the bot is able to effectively complete the given tasks and shows promise for completing more difficult tasks within complex environments. Presentation Time: Thursday, 1-2 p.m. Zoom link: https://usu-edu.zoom.us/j/87293114488?pwd=QmxrWlM0TnB1RWdPbnkvRE9WZEZIZz09

Malicious Applications Of Artificial Intelligence Escalate Existing Vulnerabilities To Cyberattacks, Cayden Groicher

Student Research Symposium

The democratization of artificial intelligence (AI) technology paired with existing malware technologies and social-engineering methods presents a dynamic threat that many existing cybersecurity systems are not prepared to mitigate. Malicious AI applied in cyberattacks is highly plausible in the near-term time-horizon and would be very characteristic of state and non-state actors. The subset of malware called ransomware is a likely candidate whose proliferation, when paired with AI, could have wide-reaching consequences to many individuals. NotPetya and WannaCry, are two examples of the immense financial damages that result from an effective ransomware attack. AI being used to instigate a ransomware attack …

Ai-Driven Drone Swarms Pose Multiple Security Threats To The Us Military, Andy Pierucci

Student Research Symposium

The United States (US) faces threats from a variety of state actors including strategic competitors (i.e. China and Russia), regional powers (i.e. Iran), weak or failed states (i.e. North Korea) and numerous non-state actors including international and domestic terrorist organizations and lone wolf radicals. As drones become more autonomous, nimbler, and less expensive they are becoming easier for state and non-state actors to manufacture or purchase and improve. Artificial intelligence (AI) driven drone swarms pose a unique threat for the US military. Advances in AI technology in the next five to ten years will expand drone capabilities, strategies, and tactics, …

China Becoming The First Artificial Intelligence Superpower Widens Tech Gap, Poses New Threats, Henry Blackburn

Student Research Symposium

President Xi Jinping declared in October 2017 “We will….speed-up development of intelligent military [AI], and improve combat capabilities for joint operations based on the network information system and the ability to fight under multi-dimensional conditions. China continued in 2017 and was among the first nations to advance in a national-AI development strategy that broadly address AI’s role in economic development.This presentation is focused on these technological developments of AI and what that means for the US. Presentation Time: Wednesday, 3-4 p.m.

Artificial Intelligence And Job Automation: Challenges And Opportunities Presented For Future Workforces, Stevie Graham

Student Research Symposium

Job Automation poses many different risks to the population. While AI can be beneficial when it comes to removing human error in some industries, it can also create new levels of concern. There are inherent risks that Artificial Intelligence (AI) presents that may result in a mass job loss worldwide in the financial industry. The finance industry faces a risk of mass job losses which could result in a hurting economy. It is estimated that by the year 2030 over 20 million jobs will be replaced by artificial intelligence. In banking and finance Artificial Intelligence is currently assisting businesses by …

“A Fitbit For Your Brain”-Elon Musk, Sci-Fi Or Attainable?, Ryan Cvelbar

Osmosis Magazine

Just a couple months ago, Elon Musk, the CEO of Tesla, SpaceX, and now Neuralink, debuted his enhanced design of the Neuralink, a wireless brain implant the size of a coin, which he plans to use to ultimately facilitate the achievement of a state of symbiosis between humans and artificial intelligence. Specifically, Musk envisions a world where we can control technology with our minds by connecting the neurons of our brains to the Neuralink and hence, digitally to computers.

A Scopus-Based Bibliometric Study Of Maritime Research Involving The Automatic Identification System, Steven D. Meyers, Laura Azevedo, Mark E. Luther

Marine Science Faculty Publications

Vessel traffic records from the Automatic Identification System (AIS) are a useful source of information for maritime data analytics, and of training data for maritime artificial intelligence systems. Researchers utilizing these data are developing the foundations for operational maritime tools essential to economic expansion and security. The global growth and distribution of this research effort from 1997 to 2019 was examined through a bibliometric study of 817 Scopus-listed publications. Indications of both collaboration and accelerating competition were found by examining the number of publications and authors, national and institutional affiliations of the authors, and number of citations received. Prior …