Explore chapters and articles related to this topic
Artificial intelligence as a feminist bioethics issue
Published in Wendy A. Rogers, Jackie Leach Scully, Stacy M. Carter, Vikki A. Entwistle, Catherine Mills, The Routledge Handbook of Feminist Bioethics, 2022
Promises of better health outcomes notwithstanding, longstanding intersecting bioethical issues persist in the expanding arena of AI health technologies. In particular, there are questions around whether the opacity of DL neural networks, non-transparent data sharing practices, and uncertain clinical value and safety of these emerging technologies may compromise people’s ability to provide truly informed consent and avoid iatrogenic harm that may result from inaccurate or biased AI analytics. From breast cancer screening to fertility and IVF prediction models, concerns abound regarding the data quality, evidence gaps, or clinical applicability for many AI-powered platforms (Peragallo Urrutia et al. 2018; Houssami et al. 2019; Wang et al. 2019; Carter et al. 2020).
Neural Networks for Medical Image Computing
Published in K. Gayathri Devi, Kishore Balasubramanian, Le Anh Ngoc, Machine Learning and Deep Learning Techniques for Medical Science, 2022
V.A. Pravina, P.K. Poonguzhali, A Kishore Kumar
The neural network contains algorithms that function in the same way as that of the human brain and helps in finding how the data are related to each other. They are inspired by the human brain, mimicking the way the human brain operates. The neural networks have numerous neurons similar to the biological neurons that transfer the signal to one another. These neural networks study the environment and adapt to it. Hence it could possibly analyze the different inputs and produce the outputs without modifying the entire structure.
A Review of Automatic Cardiac Segmentation using Deep Learning and Deformable Models
Published in Kayvan Najarian, Delaram Kahrobaei, Enrique Domínguez, Reza Soroushmehr, Artificial Intelligence in Healthcare and Medicine, 2022
Behnam Rahmatikaregar, Shahram Shirani, Zahra Keshavarz-Motamed
Neural networks are a type of supervised learning algorithm and are the basis of deep learning frameworks. A neural network includes a set of nodes (neurons) that are connected to each other via weighted links and are distributed in multiple layers. Each neural network has an input layer, several hidden layers, and an output layer. Figure 2.4 shows an example of an artificial neural network.
Advancing cervical cancer diagnosis and screening with spectroscopy and machine learning
Published in Expert Review of Molecular Diagnostics, 2023
Carlos A. Meza Ramirez, Michael Greenop, Yasser A. Almoshawah, Pierre L. Martin Hirsch, Ihtesham U. Rehman
Neural networks are computational models which can classify or organize datasets. Schematically, a neural network model consists of different processing units which are connected and run in parallel. These processing units aim to mimic a neuron from a biological brain where a series of decisions occur. A neural network may consist of an ‘n’ number of layers, each with an ‘n’ number of process units (Figure 2) [24,49,50]. The number of layers depends on the type of neural network; shallow or deep neural network. Whereas the number of processing units depends on the model input defined by the user. A shallow neural network may have between 5 and 10, but this number may differ [24,49,50]. The product of the mathematical process that occurs on each processing unit is classified by the activation function. In other words, the activation function is responsible for characterizing the data by classes, or groups [24,49,50]. Based on the author’s experience, in spectroscopy it is recommended to use ReLu, sigmoid, or ReLu activation functions, for binary or multiclass predictions, however, this is subject to the users’ criteria.
Capabilities of neural network technologies for extracting new medical knowledge and enhancing precise decision making for patients
Published in Expert Review of Precision Medicine and Drug Development, 2022
Leonid N. Yasnitsky, Andrey A. Dumler, Fedor M. Cherepanov, Vitaly L. Yasnitsky, Natalia A. Uteva
The neural network created in this traditional way is suitable for solving problems of diagnostics and classification of diseases. But it is completely unsuitable for tasks of scenario prediction, optimization, management and extraction of new scientific knowledge. For example, if, having diagnosed a patient with some disease, we want to make a scenario forecast of the disease progression for 10 years ahead by virtual increase of the patient’s age by 10 years, while keeping the other input parameters unchanged, such a forecast will be wrong. The fact is that, increasing the patient’s age, we must somehow change many other parameters – biochemical analysis data, electrocardiography, coronarography, complaints and many other age correlated parameters. But we cannot take such changes into account due to the lack of formalized knowledge about such dependencies.
Recent evolutions of machine learning applications in clinical laboratory medicine
Published in Critical Reviews in Clinical Laboratory Sciences, 2021
Sander De Bruyne, Marijn M. Speeckaert, Wim Van Biesen, Joris R. Delanghe
There has been a recent rise in various ML applications in the field of clinical laboratory medicine. Despite the potential of ML to ameliorate the efficiency of laboratory processes and optimize diagnostic workflows, translation into routine practice is still slow-going. There is a need to raise more awareness about the vast ML landscape among laboratory professionals. Educational programs dealing with theoretical ML concepts as well as their associated challenges and opportunities could stimulate wider acceptance and exploitation in the clinical laboratory. It is important to realize that ML will not immediately function as a surrogate of the laboratory professional’s neural networks, but will rather act as a valuable supportive tool with the capability of increasing the odds on optimal outcomes for patients accessing health care.