Application of Natural Language Processing in Healthcare

Roy, Khushi; Debdas, Subhra; Kundu, Sayantan; Chouhan, Shalini; Mohanty, Shivangi; Biswas, Biswarup

doi:10.1002/9781119818717.ch21

Cited by 10 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is pretty tough to get a precise orientation because of the differences or inconsistencies between the scientific estimates and discrepancies in the methodological evaluations. 7 This paper aims to do the same for the medical community, where no code AI tool can be used for free and will help healthcare professionals save…”

Section: The Physicianmentioning

confidence: 99%

Introduction to Artificial Intelligence Prediction for Healthcare

Chakravorty

2023

phy

View full text Add to dashboard Cite

Artificial intelligence (AI) aims to mimic human cognitive functions. It is bringing a paradigm shift to healthcare, powered by the increasing availability of healthcare data and rapid progress of analytics techniques.1 AI can be applied to various types of healthcare data (structured and unstructured). Popular AI techniques include machine learning methods for structured data, such as the classical support vector machine and neural network, modern deep learning, and natural language processing for unstructured data. Major disease areas that use AI tools include cancer, neurology and cardiology.

show abstract

Section: The Physicianmentioning

confidence: 99%

Introduction to Artificial Intelligence Prediction for Healthcare

Chakravorty

2023

phy

View full text Add to dashboard Cite

show abstract

“…It can allow them to have more in-person time with their patients and potentially speed up the treatment process and augment clinical decision making. AI in healthcare aims to harness the power of advanced computational techniques and algorithms to analyze and interpret extensive and complex medical datasets, consequently aiding clinical decision making [ 9 , 10 , 11 , 12 , 13 ]. Numerous studies have demonstrated the potential of AI to augment clinical procedures and patient safety [ 9 , 10 , 11 ].…”

Section: Introductionmentioning

confidence: 99%

“…AI in healthcare aims to harness the power of advanced computational techniques and algorithms to analyze and interpret extensive and complex medical datasets, consequently aiding clinical decision making [ 9 , 10 , 11 , 12 , 13 ]. Numerous studies have demonstrated the potential of AI to augment clinical procedures and patient safety [ 9 , 10 , 11 ]. However, the benefits of AI can be realized when the end user, that is, the clinician, can use it effectively (correctly) and efficiently (timely) [ 14 ].…”

Section: Introductionmentioning

confidence: 99%

Clinicians’ Perceptions of Artificial Intelligence: Focus on Workload, Risk, Trust, Clinical Decision Making, and Clinical Integration

Shamszare

Choudhury

2023

Healthcare

View full text Add to dashboard Cite

Artificial intelligence (AI) offers the potential to revolutionize healthcare, from improving diagnoses to patient safety. However, many healthcare practitioners are hesitant to adopt AI technologies fully. To understand why, this research explored clinicians’ views on AI, especially their level of trust, their concerns about potential risks, and how they believe AI might affect their day-to-day workload. We surveyed 265 healthcare professionals from various specialties in the U.S. The survey aimed to understand their perceptions and any concerns they might have about AI in their clinical practice. We further examined how these perceptions might align with three hypothetical approaches to integrating AI into healthcare: no integration, sequential (step-by-step) integration, and parallel (side-by-side with current practices) integration. The results reveal that clinicians who view AI as a workload reducer are more inclined to trust it and are more likely to use it in clinical decision making. However, those perceiving higher risks with AI are less inclined to adopt it in decision making. While the role of clinical experience was found to be statistically insignificant in influencing trust in AI and AI-driven decision making, further research might explore other potential moderating variables, such as technical aptitude, previous exposure to AI, or the specific medical specialty of the clinician. By evaluating three hypothetical scenarios of AI integration in healthcare, our study elucidates the potential pitfalls of sequential AI integration and the comparative advantages of parallel integration. In conclusion, this study underscores the necessity of strategic AI integration into healthcare. AI should be perceived as a supportive tool rather than an intrusive entity, augmenting the clinicians’ skills and facilitating their workflow rather than disrupting it. As we move towards an increasingly digitized future in healthcare, comprehending the among AI technology, clinician perception, trust, and decision making is fundamental.

show abstract

“…In an era of data explosion, there is an increasing demand for various fields to launch AI-driven applications in image classification [1], natural language processing (NLP) [2], and so forth. Behind these applications are numerous models that have been fit in hugesize datasets such as ImageNet [3].…”

Section: Introductionmentioning

confidence: 99%

Sign Bit is Enough: A Learning Synchronization Framework for Multi-hop All-reduce with Ultimate Compression

Wu¹,

He²,

Guo³

et al. 2022

Preprint

View full text Add to dashboard Cite

Traditional one-bit compressed stochastic gradient descent can not be directly employed in multi-hop all-reduce, a widely adopted distributed training paradigm in network-intensive high-performance computing systems such as public clouds. According to our theoretical findings, due to the cascading compression, the training process has considerable deterioration on the convergence performance. To overcome this limitation, we implement a sign-bit compression-based learning synchronization framework, Marsit. It prevents cascading compression via an elaborate bit-wise operation for unbiased sign aggregation and its specific global compensation mechanism for mitigating compression deviation. The proposed framework retains the same theoretical convergence rate as noncompression mechanisms. Experimental results demonstrate that Marsit reduces up to 35% training time while preserving the same accuracy as training without compression.

show abstract

Application of Natural Language Processing in Healthcare

Cited by 10 publications

References 15 publications

Introduction to Artificial Intelligence Prediction for Healthcare

Introduction to Artificial Intelligence Prediction for Healthcare

Clinicians’ Perceptions of Artificial Intelligence: Focus on Workload, Risk, Trust, Clinical Decision Making, and Clinical Integration

Sign Bit is Enough: A Learning Synchronization Framework for Multi-hop All-reduce with Ultimate Compression

Contact Info

Product

Resources

About