Components of visual assessment in the diagnosis of effusions were analyzed using relative operating characteristic. Diagnostic performance in the assessment of malignancy and the specification of metastatic origin was measured for two expert cytologists. The component of performance attributable to feature interpretation was measured in protocols which minimized the effects of clinical information and visual search in the decision process. Feature interpretation, as a process, contributed significantly to the evaluation of malignancy and marginally to the specification of metastatic origin. For each of these diagnostic tasks, the process of feature interpretation was codified in the construction of explicit models. The expert cytologists were asked to define a set of localized visual features that incorporate essential visual elements for diagnosis. These features were evaluated for a set of test cases, and regression models were constructed defining malignancy and metastatic origin. Relative operating characteristic analysis indicated that the predictive value of the models for diagnosis was very similar to the component of human performance attributable to feature interpretation.

Download full-text PDF

Source

Publication Analysis

Top Keywords

feature interpretation
16
metastatic origin
12
visual features
8
relative operating
8
operating characteristic
8
specification metastatic
8
expert cytologists
8
performance attributable
8
attributable feature
8
process feature
8

Similar Publications

The transportation industry contributes significantly to climate change through carbon dioxide ( ) emissions, intensifying global warming and leading to more frequent and severe weather phenomena such as flooding, drought, heat waves, glacier melting, and rising sea levels. This study proposes a comprehensive approach for predicting emissions from vehicles using deep learning techniques enhanced by eXplainable Artificial Intelligence (XAI) methods. Utilizing a dataset from the Canadian government's official open data portal, we explored the impact of various vehicle attributes on emissions.

View Article and Find Full Text PDF

AI integration into wavelength-based SPR biosensing: Advancements in spectroscopic analysis and detection.

Anal Chim Acta

March 2025

Artificial Intelligence Research Center, Chang Gung University, Taoyuan, 333323, Taiwan; Department of Artificial Intelligence, College of Intelligent Computing, Chang Gung University, Taoyuan, 333323, Taiwan. Electronic address:

Background: In recent years, employing deep learning methods in the biosensing area has significantly reduced data analysis time and enhanced data interpretation and prediction accuracy. In some SPR fields, research teams have further enhanced detection capabilities using deep learning techniques. However, the application of deep learning to spectroscopic surface plasmon resonance (SPR) biosensors has not been reported.

View Article and Find Full Text PDF

Comparative assessment of microplastic ingestion among deep sea decapods: distribution analysis in Sardinian and Catalan waters.

Environ Res

January 2025

Dipartimento di Scienze della Vita e dell'Ambiente, Università degli Studi di Cagliari, Via Tommaso Fiorelli 1, 09126 Cagliari, Italy; Departament de Biologia Animal, Biologia Vegetal i Ecologia, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain. Electronic address:

Microplastic (MP) pollution is a widespread and concerning environmental issue. The benthic layer is known as one of the major accumulation sinks, yet knowledge gaps still remain in describing the interactions of its biota with MPs. This work represents a comprehensive comparative analysis of MP ingestion in the four deep-sea crustacean decapods Aristeus antennatus (Risso, 1816), Aristaeomorpha foliacea (Risso, 1827), Nephrops norvegicus (Linnaeus, 1758) and Parapenaeus longirostris (Lucas, 1846) sampled from two distinct regions of the Mediterranean Sea in order to underscore the species-specific characteristics driving their MP ingestion variations.

View Article and Find Full Text PDF

Beyond XGBoost and SHAP: Unveiling true feature importance.

J Hazard Mater

January 2025

Faculty of Data Science, Musashino University, 3-3-3 Ariake Koto-ku, Tokyo 135-8181, Japan. Electronic address:

This paper outlines key machine learning principles, focusing on the use of XGBoost and SHAP values to assist researchers in avoiding analytical pitfalls. XGBoost builds models by incrementally adding decision trees, each addressing the errors of the previous one, which can result in inflated feature importance scores due to the method's emphasis on misclassified examples. While SHAP values provide a theoretically robust way to interpret predictions, their dependence on model structure and feature interactions can introduce biases.

View Article and Find Full Text PDF

Shallow whole-genome sequencing (sWGS) offers a cost-effective approach to detect copy number alterations (CNAs). However, there remains a gap for a standardized workflow specifically designed for sWGS analysis. To address this need, in this work we present SAMURAI, a bioinformatics pipeline specifically designed for analyzing CNAs from sWGS data in a standardized and reproducible manner.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!