A hyperspectral dataset of precancerous lesions in gastric cancer and benchmarks for pathological diagnosis

Histological Hyperspectral Glioblastoma Dataset (HistologyHSI-GB)

Hyperspectral (HS) imaging (HSI) technology combines the main features of two existing technologies: imaging and spectroscopy. This allows to analyse simultaneously the morphological and chemical attributes of the objects captured by a HS camera. In recent years, the use of HSI provides valuable insights into the interaction between light and biological tissues, and makes it possible to detect patterns, cells, or biomarkers, thus, being able to identify diseases. This work presents the HistologyHSI-GB dataset, which contains 469 HS images from 13 patients diagnosed with brain tumours, specifically glioblastoma. The slides were stained with haematoxylin and eosin (H&E) and captured using a microscope at 20× power magnification. Skilled histopathologists diagnosed the slides and provided image-level annotations. The dataset was acquired using custom HSI instrumentation, consisting of a microscope equipped with an HS camera covering the spectral range from 400 to 1000 nm.

Targeted plasma proteomic analysis uncovers a high-performance biomarker panel for early diagnosis of gastric cancer

BACKGROUND

Gastric cancer (GC) is characterized by high morbidity, high mortality and low early diagnosis rate. Early diagnosis plays a crucial role in radically treating GC. The aim of this study was to identify plasma biomarkers for GC and early GC diagnosis.

METHODS

We quantified 369 protein levels with plasma samples from discovery cohort (n = 88) and validation cohort (n = 50) via high-throughput proximity extension assay (PEA) utilizing the Olink-Explore-384-Cardiometabolic panel. The multi-protein signatures were derived from LASSO and Ridge regression models.

RESULTS

In the discovery cohort, 13 proteins (GDF15, ITIH3, BOC, DPP7, EGFR, AMY2A, CCDC80, CD163, GPNMB, LTBP2, CTSZ, CCL18 and NECTIN2) were identified to distinguish GC (Stage I-IV) and early GC (HGIN-I) groups from control group with AUC of 0.994 and AUC of 0.998, severally. The validation cohort yielded AUC of 0.930 and AUC of 0.818 for GC and early GC, respectively.

CONCLUSIONS

This study identified a multi-protein signature with the potential to benefit clinical GC diagnosis, especially for Asian and early GC patients, which may contribute to the development of a less-invasive, convenient, and efficient early screening tool, promoting early diagnosis and treatment of GC and ultimately improving patient survival.

Rapid and robust bacterial species identification using hyperspectral microscopy and gram staining techniques

Gram staining can classify bacterial species into two large groups based on cell wall differences. Our study revealed that within the same gram group (gram-positive or gram-negative), subtle cell wall variations can alter staining outcomes, with the peptidoglycan layer and lipid content significantly influencing this effect. Thus, bacteria within the same group can also be differentiated by their spectra. Using hyperspectral microscopy, we identified six species of intestinal bacteria with 98.1% accuracy. Our study also demonstrated that selecting the right spectral band and background calibration can enhance the model's robustness and facilitate precise identification of varying sample batches. This method is suitable for analyzing bacterial community pathologies.

CrossU-Net: Dual-modality cross-attention U-Net for segmentation of precancerous lesions in gastric cancer

Gastric precancerous lesions (GPL) significantly elevate the risk of gastric cancer, and precise diagnosis and timely intervention are critical for patient survival. Due to the elusive pathological features of precancerous lesions, the early detection rate is less than 10%, which hinders lesion localization and diagnosis. In this paper, we provide a GPL pathological dataset and propose a novel method for improving the segmentation accuracy on a limited-scale dataset, namely RGB and Hyperspectral dual-modal pathological image Cross-attention U-Net (CrossU-Net). Specifically, we present a self-supervised pre-training model for hyperspectral images to serve downstream segmentation tasks. Secondly, we design a dual-stream U-Net-based network to extract features from different modal images. To promote information exchange between spatial information in RGB images and spectral information in hyperspectral images, we customize the cross-attention mechanism between the two networks. Furthermore, we use an intermediate agent in this mechanism to improve computational efficiency. Finally, we add a distillation loss to align predicted results for both branches, improving network generalization. Experimental results show that our CrossU-Net achieves accuracy and Dice of 96.53% and 91.62%, respectively, for GPL lesion segmentation, providing a promising spectral research approach for the localization and subsequent quantitative analysis of pathological features in early diagnosis.

A novel label-free capillary electrophoresis LED-induced fluorescence platform based on catalytic hairpin assembly for sensitive detection of multiple circulating tumor DNA

Circulating tumor DNA (ctDNA) is a highly promising biomarker for the early diagnosis and treatment of gastric cancer (GC). However, there is still a lack of effective and practical ctDNA detection methods. In this work, a simple and economical capillary non-gel sieving electrophoresis-LED induced fluorescence detection (NGCE-LEDIF) platform coupled with catalytic hairpin assembly (CHA) as the signal amplification strategy is proposed for quantitative detection of PIK3CA E542K and TP53 (two types of ctDNA associated with GC). We have reasonably designed two pairs of programmable oligonucleotide hairpin probes for PIK3CA E542K and TP53. Using a one-pot reaction, the presence of ctDNA triggers the cyclic amplification of CHA, forming numerous thermodynamically stable H1/H2 double-strands. The H1/H2 double-stranded DNA catalyzed by PIK3CA E542K and TP53 can be easily separated by NGCE due to their different lengths, enabling simultaneous detection of both ctDNAs. Under optimal experimental conditions, the detection limits of this strategy for detecting GC-related biomarkers PIK3CA E542K and TP53 are 20.35 pM and 19.61 pM, respectively, and can achieve 730-fold signal amplification. This strategy has a good recovery in the serum matrix. The results of this study show that this strategy has significant advantages such as high selectivity, a simple process, no special instruments and equipment, no need for fluorescence modification of hairpin probes in advance, high automation, low cost, and minimal sample consumption. This provides a powerful method for the detection of trace cancer biomarkers in the serum matrix with good application prospects.

A framework for immunofluorescence image augmentation and classification based on unsupervised attention mechanism

Autoimmune encephalitis (AE) is a common neurological disorder. As a standard method for neuroautoantibody detection, pathologists use tissue matrix assays (TBA) for initial disease screening. In this study, microscopic fluorescence imaging was combined with deep learning to improve AE diagnostic accuracy. Due to the inter-class imbalance of medical data, we propose an innovative generative adversarial network supplemented with attention mechanisms to highlight key regions in images to synthesize high-quality fluorescence images. However, securing annotated medical data is both time-consuming and costly. To circumvent this problem, we employ a self-supervised learning approach that utilizes unlabeled fluorescence data to support downstream classification tasks. To better understand the fluorescence properties in the data, we introduce a multichannel input convolutional neural network that adds additional channels of fluorescence intensity. This study builds an AE immunofluorescence dataset and obtains the classification accuracy of 88.5% using our method, thus confirming the effectiveness of the proposed method.

Joint Diagnostic Method of Tumor Tissue Based on Hyperspectral Spectral-Spatial Transfer Features

In order to improve the clinical application of hyperspectral technology in the pathological diagnosis of tumor tissue, a joint diagnostic method based on spectral-spatial transfer features was established by simulating the actual clinical diagnosis process and combining micro-hyperspectral imaging with large-scale pathological data. In view of the limited sample volume of medical hyperspectral data, a multi-data transfer model pre-trained on conventional pathology datasets was applied to the classification task of micro-hyperspectral images, to explore the differences in spectral-spatial transfer features in the wavelength of 410-900 nm between tumor tissues and normal tissues. The experimental results show that the spectral-spatial transfer convolutional neural network (SST-CNN) achieved a classification accuracy of 95.46% for the gastric cancer dataset and 95.89% for the thyroid cancer dataset, thus outperforming models trained on single conventional digital pathology and single hyperspectral data. The joint diagnostic method established based on SST-CNN can complete the interpretation of a section of data in 3 min, thus providing a new technical solution for the rapid diagnosis of pathology. This study also explored problems involving the correlation between tumor tissues and typical spectral-spatial features, as well as the efficient transformation of conventional pathological and transfer spectral-spatial features, which solidified the theoretical research on hyperspectral pathological diagnosis.