Are Artificial Intelligence-Assisted Three-Dimensional Histological Reconstructions Reliable for the Assessment of Trabecular Microarchitecture?

Objectives: This study aimed to create a three-dimensional histological reconstruction through the AI-assisted classification of tissues and the alignment of serial sections. The secondary aim was to evaluate if the novel technique for histological reconstruction accurately replicated the trabecular microarchitecture of bone. This was performed by conducting micromorphometric measurements on the reconstruction and comparing the results obtained with those of microCT reconstructions. Methods: A bone biopsy sample was harvested upon re-entry following sinus floor augmentation. Following microCT scanning and histological processing, a modified version of the U-Net architecture was trained to categorize tissues on the sections. Detector-free local feature matching with transformers was used to create the histological reconstruction. The micromorphometric parameters were calculated using Bruker's CTAn software (version 1.18.8.0, Bruker, Kontich, Belgium) for both histological and microCT datasets. Results: Correlation coefficients calculated between the micromorphometric parameters measured on the microCT and histological reconstruction suggest a strong linear relationship between the two with p-values of 0.777, 0.717, 0.705, 0.666, and 0.687 for BV/TV, BS/TV, Tb.Pf Tb.Th, and Tb.Sp, respectively. Bland-Altman and mountain plots suggest good agreement between BV/TV measurements on the two reconstruction methods. Conclusions: This novel method for three-dimensional histological reconstruction provides researchers with a tool that enables the assessment of accurate trabecular microarchitecture and histological information simultaneously.

HunCRC: annotated pathological slides to enhance deep learning applications in colorectal cancer screening.

Histopathology is the gold standard method for staging and grading human tumors and provides critical information for the oncoteam's decision making. Highly-trained pathologists are needed for careful microscopic analysis of the slides produced from tissue taken from biopsy. This is a time-consuming process. A reliable decision support system would assist healthcare systems that often suffer from a shortage of pathologists. Recent advances in digital pathology allow for high-resolution digitalization of pathological slides. Digital slide scanners combined with modern computer vision models, such as convolutional neural networks, can help pathologists in their everyday work, resulting in shortened diagnosis times. In this study, 200 digital whole-slide images are published which were collected via hematoxylin-eosin stained colorectal biopsy. Alongside the whole-slide images, detailed region level annotations are also provided for ten relevant pathological classes. The 200 digital slides, after pre-processing, resulted in 101,389 patches. A single patch is a 512 × 512 pixel image, covering 248 × 248 µm2 tissue area. Versions at higher resolution are available as well. Hopefully, HunCRC, this widely accessible dataset will aid future colorectal cancer computer-aided diagnosis and research.

Correlation between Expression Profiles of Key Signaling Genes in Colorectal Cancer Samples from Type 2 Diabetic and Non-Diabetic Patients.

Several lines of epidemiological and biochemical evidence support the association of type 2 diabetes mellitus (T2DM) and colorectal cancer (CRC). T2DM has been shown to impinge on the transcriptome of colon tumor cells, promoting their proliferation and invasion. In order to gain insight into diabetes-specific modulation of colon cancer signaling, we analyzed gene expression patterns of more than five hundred genes encoding signaling proteins on TaqMan OpenArray panels from colonoscopic colorectal tumor samples of type 2 diabetic and non-diabetic patients. In total, 48 transcripts were found to be differentially expressed in tumors of T2DM patients as compared to healthy colon samples. Enrichment analysis with the g:GOSt (Gene Ontology Statistics) functional profiling tool revealed that the underlying genes can be classified into five signaling pathways (in decreasing order of significance: Wnt (wingless-type)/ß-catenin; Hippo; TNF (tumor necrosis factor); PI3K/Akt (phosphoinositide-3 kinase/protein kinase B), and platelet activation), implying that targeted downregulation of these signaling cascades might help combat CRC in diabetic patients. Transcript levels of some of the differentially expressed genes were also measured from surgically removed diabetic and non-diabetic CRC specimens by individual qPCR (quantitative real-time PCR) assays using the adjacent normal tissue mRNA levels as an internal control. The most significantly altered genes in diabetic tumor samples were largely different from those in non-diabetic ones, implying that T2DM profoundly alters the expression of signaling genes and presumably the biological characteristics of CRC.

The Effect of Methylselenocysteine and Sodium Selenite Treatment on microRNA Expression in Liver Cancer Cell Lines.

The unique character of selenium compounds, including sodium selenite and Se-methylselenocysteine (MSC), is that they exert cytotoxic effects on neoplastic cells, providing a great potential for treating cancer cells being highly resistant to cytostatic drugs. However, selenium treatment may affect microRNA (miRNA) expression as the pattern of circulating miRNAs changed in a placebo-controlled selenium supplement study. This necessitates exploring possible changes in the expression profiles of miRNAs. For this, miRNAs being critical for liver function were selected and their expression was measured in hepatocellular carcinoma (HLE and HLF) and cholangiocarcinoma cell lines (TFK-1 and HuH-28) using individual TaqMan MicroRNA Assays following selenite or MSC treatments. For establishing tolerable concentrations, IC50 values were determined by performing SRB proliferation assays. The results revealed much lower IC50 values for selenite (from 2.7 to 11.3 µM) compared to MSC (from 79.5 to 322.6 µM). The treatments resulted in cell line-dependent miRNA expression patterns, with all miRNAs found to show fold change differences; however, only a few of these changes were statistically different in treated cells compared to untreated cells below IC50. Namely, miR-199a in HLF, miR-143 in TFK-1 upon MSC treatment, miR-210 in HLF and TFK-1, miR-22, -24, -122, -143 in HLF upon selenite treatment. Fold change differences revealed that miR-122 with both selenium compounds, miR-199a with MSC and miR-22 with selenite were affected. The miRNAs showing minimal alterations included miR-125b and miR-194. In conclusion, our results revealed moderately altered miRNA expression in the cell lines (less alterations following MSC treatment), being miR-122, -199a the most affected and miR-125b, -194 the least altered miRNAs upon selenium treatment.

Cholangiocarcinoma: Classification, Histopathology and Molecular Carcinogenesis.

Cholangiocarcinoma (CC) is the second most common tumor of the liver, originating from the biliary system with increasing incidence and mortality worldwide. Several new classifications review the significance of tumor localization, site of origin, proliferation and biomarkers in the intrahepatic, perihilar and distal forms of the lesion. Based on growth pattern mass-forming, periductal-infiltrating, intraductal, undefined and mixed types are differentiated. There are further subclassifications which are applied for the histological features, in particular for intrahepatic CC. Recognition of the precursors and early lesions of CC including biliary intraepithelial neoplasia (BilIN), intraductal papillary neoplasm of the bile ducts (IPNB), biliary mucinous cystic neoplasm (MCNB) and the candidate precursors, such as bile duct adenoma and von Meyenburg complex is of increasing significance. In addition to the previously used biliary markers detected by immunohistochemistry, several new markers have been added to the differentiation of both the benign and malignant lesions, which can be used to aid in the subclassification in association with the outcome of CC. Major aspects of biliary carcinogenesis have been revealed, yet, the exact way of this diverse process is still unclear. The factors contributing to molecular cholangiocarcinogenesis include various risk factors, different anatomical localizations, multiple cellular origins, genetic and epigenetic alterations, tumor microenvironment, heterogeneity and clonal evolution. Driver mutations have been identified, implying that they are optimal candidates for targeted therapy. The most promising therapeutic candidates have entered clinical trials.

MicroRNA Expression in Focal Nodular Hyperplasia in Comparison with Cirrhosis and Hepatocellular Carcinoma.

The liver disease focal nodular hyperplasia (FNH) has several histological features that resemble hepatic cirrhosis. Since cirrhosis may develop further into hepatocellular carcinoma (HCC) contrary to FNH, the aim of the present study was to identify microRNAs (miRNA), which, by their altered expression levels, may be associated with the benign, tumor-like nature of FNH. Altogether 106 surgically removed formalin-fixed paraffin-embedded liver samples were selected, including 22 FNH, 45 cirrhosis, 24 HCC and 15 normal liver tissues. Etiology of the cases of cirrhosis and HCC includes hepatitis C and alcoholism and the HCC cases developed in cirrhotic livers. Relative expression levels of 14 miRNAs were determined using TaqMan MicroRNA Assays. In comparison to normal liver, the levels of miR-34a and miR-224 were elevated not only in FNH but also in cirrhosis and HCC, while the expression of miR-17-5p, miR-18a and miR-210 was decreased in FNH. Further, the levels of miR-21 and miR-222 were increased in cirrhosis and HCC but were decreased in FNH and the expression of miR-17-5p, miR-18a, miR-195 and miR-210 was decreased in FNH as compared with cirrhosis and/or HCC. In conclusion, the elevation of miR-34a and miR-224 may be associated with both benign and malignant proliferative processes, nevertheless the increased expression of oncomiRs miR-21 and miR-222 in cirrhosis and HCC but not in FNH may be related to malignant processes of the liver. The decreased levels of miR-18a, miR-195 and miR-210 may further differentiate FNH from cirrhosis, reflecting the different pathogenesis of these two entities contrary to some histologically similar features.