Your browser doesn't support javascript.
loading
Show: 20 | 50 | 100
Results 1 - 3 de 3
Filter
Add more filters










Database
Language
Publication year range
1.
Lab Invest ; 104(5): 100341, 2024 May.
Article in English | MEDLINE | ID: mdl-38280634

ABSTRACT

Ki-67 is a nuclear protein associated with proliferation, and a strong potential biomarker in breast cancer, but is not routinely measured in current clinical management owing to a lack of standardization. Digital image analysis (DIA) is a promising technology that could allow high-throughput analysis and standardization. There is a dearth of data on the clinical reliability as well as intra- and interalgorithmic variability of different DIA methods. In this study, we scored and compared a set of breast cancer cases in which manually counted Ki-67 has already been demonstrated to have prognostic value (n = 278) to 5 DIA methods, namely Aperio ePathology (Lieca Biosystems), Definiens Tissue Studio (Definiens AG), Qupath, an unsupervised immunohistochemical color histogram algorithm, and a deep-learning pipeline piNET. The piNET system achieved high agreement (interclass correlation coefficient: 0.850) and correlation (R = 0.85) with the reference score. The Qupath algorithm exhibited a high degree of reproducibility among all rater instances (interclass correlation coefficient: 0.889). Although piNET performed well against absolute manual counts, none of the tested DIA methods classified common Ki-67 cutoffs with high agreement or reached the clinically relevant Cohen's κ of at least 0.8. The highest agreement achieved was a Cohen's κ statistic of 0.73 for cutoffs 20% and 25% by the piNET system. The main contributors to interalgorithmic variation and poor cutoff characterization included heterogeneous tumor biology, varying algorithm implementation, and setting assignments. It appears that image segmentation is the primary explanation for semiautomated intra-algorithmic variation, which involves significant manual intervention to correct. Automated pipelines, such as piNET, may be crucial in developing robust and reproducible unbiased DIA approaches to accurately quantify Ki-67 for clinical diagnosis in the future.


Subject(s)
Breast Neoplasms , Image Processing, Computer-Assisted , Ki-67 Antigen , Humans , Ki-67 Antigen/analysis , Ki-67 Antigen/metabolism , Breast Neoplasms/diagnostic imaging , Breast Neoplasms/pathology , Female , Reproducibility of Results , Image Processing, Computer-Assisted/methods , Biomarkers, Tumor/metabolism , Biomarkers, Tumor/analysis , Algorithms , Immunohistochemistry/methods
2.
Sci Rep ; 13(1): 13253, 2023 08 15.
Article in English | MEDLINE | ID: mdl-37582966

ABSTRACT

In this paper, a novel technique for detecting female breast anomalous tissues is presented and validated through numerical simulations. The technique, to a high degree, resembles X-ray mammography; however, instead of using X-rays for obtaining images of the breast, low-frequency electromagnetic fields are leveraged. To capture breast impressions, a metasurface, which can be thought of as analogous to X-rays film, has been employed. To achieve deep and sufficient penetration within the breast tissues, the source of excitation is a simple narrow-band dipole antenna operating at 200 MHz. The metasurface is designed to operate at the same frequency. The detection mechanism is based on comparing the impressions obtained from the breast under examination to the reference case (healthy breasts) using machine learning techniques. Using this system, not only would it be possible to detect tumors (benign or malignant), but one can also determine the location and size of the tumors. Remarkably, deep learning models were found to achieve very high classification accuracy.


Subject(s)
Deep Learning , Female , Humans , Electromagnetic Fields , Mammography/methods , Breast/diagnostic imaging , Breast/pathology , Machine Learning
3.
Sensors (Basel) ; 19(21)2019 Nov 02.
Article in English | MEDLINE | ID: mdl-31684027

ABSTRACT

This paper presents a concept of an intelligent sensing technique based on modulating the frequency responses of microwave near-field sensors to characterize material parameters. The concept is based on the assumption that the physical parameters being extracted such as fluid concentration are constant over the range of frequency of the sensor. The modulation of the frequency response is based on the interactions between the material under test and multiple sensors. The concept is based on observing the responses of the sensors over a frequency wideband as vectors of many dimensions. The dimensions are then considered as the features for a neural network. With small datasets, the neural networks can produce highly accurate and generalized models. The concept is demonstrated by designing a microwave sensing system based on a two-port microstrip line exciting three-identical planar resonators. For experimental validation, the sensor is used to detect the concentration of a fluid material composed of two pure fluids. Very high accuracy is achieved.

SELECTION OF CITATIONS
SEARCH DETAIL
...