The Potential Dangers of Artificial Intelligence for Radiology and Radiologists

Chu, Linda C.; Anandkumar, Animashree; Shin, Hoo Chang; Fishman, Elliot K.

doi:10.1016/j.jacr.2020.04.010

Cited by 14 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At 25 mAs-TFI, there was no difference between noise and streak artifacts, but the score of small vessels was better than that of 175 mAs-FBP. It is noted that deep learning reconstruction has the risk of creating imaginary objects 17 . In this study, no structures suspected of being created by TFI were observed.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

The evaluation of the reduction of radiation dose via deep learning-based reconstruction for cadaveric human lung CT images

Miyata

Yanagawa

Kikuchi

et al. 2022

Sci Rep

View full text Add to dashboard Cite

To compare the quality of CT images of the lung reconstructed using deep learning-based reconstruction (True Fidelity Image: TFI ™; GE Healthcare) to filtered back projection (FBP), and to determine the minimum tube current–time product in TFI without compromising image quality. Four cadaveric human lungs were scanned on CT at 120 kVp and different tube current–time products (10, 25, 50, 75, 100, and 175 mAs) and reconstructed with TFI and FBP. Two image evaluations were performed by three independent radiologists. In the first experiment, using the same tube current–time product, a side-by-side TFI and FBP comparison was performed. Images were evaluated with regard to noise, streak artifacts, and overall image quality. Overall image quality was evaluated in view of whole image quality. In the second experiment, CT images reconstructed using TFI and FBP with five different tube current–time products were displayed in random order, which were evaluated with reference to the 175 mAs-FBP image. Images were scored with regard to normal structure, abnormal findings, noise, streak artifacts, and overall image quality. Median scores from three radiologists were statistically analyzed. Quantitative evaluation of noise was performed by setting regions of interest (ROIs) in air. In first experiment, overall image quality was improved, and noise was decreased in images of TFI compared to that of FBP for all tube current–time products. In second experiment, scores of all evaluation items except for small vessels in images of 25 mAs-TFI were almost the same as that of 175 mAs-FBP (all p > 0.31). Using TFI instead of FBP, at least 85% radiation dose reduction could be possible without any degradation in the image quality.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Using noise reduction by deep learning algorithms can sharpen edges and make image clear 21 – 23 . However, we need to be careful because it may erase real structures or depict structures that do not exist 17 .…”

Section: Discussionmentioning

confidence: 99%

The evaluation of the reduction of radiation dose via deep learning-based reconstruction for cadaveric human lung CT images

Miyata

Yanagawa

Kikuchi

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…These data Fig. 7 Major key pros of Internet of things for combating coronavirus pandemic are also helpful in implementing appropriate procedures [21]. Various IoT applications:…”

Section: Iot Applications To Fight Covid-19mentioning

confidence: 99%

Impact of Artificial Intelligence and Internet of Things in Effective Handling of Coronavirus Crisis

Jaju

Thakkar

2021

Impact of AI and Data Science in Response to Coronavirus Pandemic

View full text Add to dashboard Cite

Artificial intelligence (AI) techniques integrated with real-time data collection, wireless set-ups and performance of end-user devices are now in high demand. Presently, it is in great demand to use AI to track and predict major natural pandemics [1]. The coronavirus 2019 pandemic (COVID-19), from Wuhan, has impacted a devastating effect on the community worldwide and has placed a heavy burden on high-quality health systems around the world.Today, when the whole world is struggling to survive, artificial intelligence comes into play with everyday new ways coming up to help us fight the pandemic. There are AI-powered predictive systems, advanced warning systems, AI driven diagnosis, robots, drones and more [2].Smart AI-powered platforms help us keep track of the spread of the disease and monitor it. With such platforms, it creates data structure that makes it easy to track the spread of disease, keep updated information, get immediate help and reports, and much more. BlueDot is one such platform developed to trace, detect and understand the spread of contagious diseases [3]. It achieves this by using machine learning based advanced algorithms that analyze data from number of sources, including a health organization, population and veterinary fitness reports.The Indian government is using the Arogya Setu app (Fig. 1) to help boost us the fight against COVID-19 pandemic. The application keeps you updated with day to day number of infected active cases, recovered cases and death cases. It also K. Jaju (&)

show abstract

“…Therefore, policy decisions, as well as patient monitoring and diagnosis, could be biased or wrongly conducted because of the wrong input. Therefore, a significant amount of risk is associated during the automation of the AI and ML-based systems [57]. GAN, based on a class of deep learning and machine learning class, can create new images or datasets with the same statistics as the training dataset.…”

Section: Potential Risks Of Ai and MLmentioning

confidence: 99%

Building Resilience against COVID-19 Pandemic Using Artificial Intelligence, Machine Learning, and IoT: A Survey of Recent Progress

Abir¹,

Islam

Mahmood

et al. 2020

IoT

View full text Add to dashboard Cite

Coronavirus disease 2019 (COVID-19) has significantly impacted the entire world today and stalled off regular human activities in such an unprecedented way that it will have an unforgettable footprint on the history of mankind. Different countries have adopted numerous measures to build resilience against this life-threatening disease. However, the highly contagious nature of this pandemic has challenged the traditional healthcare and treatment practices. Thus, artificial intelligence (AI) and machine learning (ML) open up new mechanisms for effective healthcare during this pandemic. AI and ML can be useful for medicine development, designing efficient diagnosis strategies and producing predictions of the disease spread. These applications are highly dependent on real-time monitoring of the patients and effective coordination of the information, where the Internet of Things (IoT) plays a key role. IoT can also help with applications such as automated drug delivery, responding to patient queries, and tracking the causes of disease spread. This paper represents a comprehensive analysis of the potential AI, ML, and IoT technologies for defending against the COVID-19 pandemic. The existing and potential applications of AI, ML, and IoT, along with a detailed analysis of the enabling tools and techniques are outlined. A critical discussion on the risks and limitations of the aforementioned technologies are also included.

show abstract

The Potential Dangers of Artificial Intelligence for Radiology and Radiologists

Cited by 14 publications

References 5 publications

The evaluation of the reduction of radiation dose via deep learning-based reconstruction for cadaveric human lung CT images

The evaluation of the reduction of radiation dose via deep learning-based reconstruction for cadaveric human lung CT images

Impact of Artificial Intelligence and Internet of Things in Effective Handling of Coronavirus Crisis

Building Resilience against COVID-19 Pandemic Using Artificial Intelligence, Machine Learning, and IoT: A Survey of Recent Progress

Contact Info

Product

Resources

About