Machine-learning-based diagnosis of thyroid fine-needle aspiration biopsy synergistically by Papanicolaou staining and refractive index distribution

Research progress on artificial intelligence technology-assisted diagnosis of thyroid diseases

With the rapid development of the "Internet + Medical" model, artificial intelligence technology has been widely used in the analysis of medical images. Among them, the technology of using deep learning algorithms to identify features of ultrasound and pathological images and realize intelligent diagnosis of diseases has entered the clinical verification stage. This study is based on the application research of artificial intelligence technology in medical diagnosis and reviews the early screening and diagnosis of thyroid diseases. The cure rate of thyroid disease is high in the early stage, but once it deteriorates into thyroid cancer, the risk of death and treatment costs of the patient increase. At present, the early diagnosis of the disease still depends on the examination equipment and the clinical experience of doctors, and there is a certain misdiagnosis rate. Based on the above background, it is particularly important to explore a technology that can achieve objective screening of thyroid lesions in the early stages. This paper provides a comprehensive review of recent research on the early diagnosis of thyroid diseases using artificial intelligence technology. It integrates the findings of multiple studies and that traditional machine learning algorithms are widely used as research objects. The convolutional neural network model has a high recognition accuracy for thyroid nodules and thyroid pathological cell lesions. U-Net network model can significantly improve the recognition accuracy of thyroid nodule ultrasound images when used as a segmentation algorithm. This article focuses on reviewing the intelligent recognition technology of thyroid ultrasound images and pathological sections, hoping to provide researchers with research ideas and help clinicians achieve intelligent early screening of thyroid cancer.

In Vivo Time-Resolved Fluorescence Detection of Liver Cancer Supported by Machine Learning

OBJECTIVES

One of the widely used optical biopsy methods for monitoring cellular and tissue metabolism is time-resolved fluorescence. The use of this method in optical liver biopsy has a high potential for studying the shift in energy-type production from oxidative phosphorylation to glycolysis and changes in the antioxidant defense of malignant cells. On the other hand, machine learning methods have proven to be an excellent solution to classification problems in medical practice, including biomedical optics. We aim to combine time-resolved fluorescence measurements and machine learning to automate the division of liver parenchyma and tumors (primary malignant, metastases and benign tumors) into classes.

MATERIALS AND METHODS

An optical biopsy was performed using a developed setup with a fine-needle optical probe in clinical conditions under ultrasound control. Fluorescence decays were recorded in a conditionally healthy liver and lesions during percutaneous needle biopsy. The labeled data set was created on the basis of the recorded fluorescence results and the histopathological classification of the biopsies obtained. Several machine learning methods were trained using different separation strategies of the training test set, and their respective accuracy was compared.

RESULTS

Our results show that each of the tumor types had its own characteristic metabolic shifts recorded by the time-resolved fluorescence spectroscopy. The application of machine learning demonstrates a reliable separation of the liver and all tumor types into cancer and noncancer classes with sensitivity, specificity and corresponding accuracy greater than 0.91, 0.79 and 0.90, using the random forest method. We also show that our method is capable of giving a preliminary diagnosis of the type of liver tumor (primary malignant, metastases and benign tumors) with a sensitivity, specificity and accuracy of at least 0.80, 0.95 and 0.90.

CONCLUSIONS

These promising results highlight its potential as a key tool in the future development of diagnostic and therapeutic strategies for liver cancers. Lasers Surg. Med. 00:00-00, 2024. 2024 Wiley Periodicals LLC.

Fourier space aberration correction for high resolution refractive index imaging using incoherent light

An aberration correction method is introduced for 3D phase deconvolution microscopy. Our technique capitalizes on multiple illumination patterns to iteratively extract Fourier space aberrations, utilizing the overlapping information inherent in these patterns. By refining the point spread function based on the retrieved aberration data, we significantly improve the precision of refractive index deconvolution. We validate the effectiveness of our method on both synthetic and biological three-dimensional samples, achieving notable enhancements in resolution and measurement accuracy. The method's reliability in aberration retrieval is further confirmed through controlled experiments with intentionally induced spherical aberrations, underscoring its potential for wide-ranging applications in microscopy and biomedicine.