Automated Nuclear Morphometry: A Deep Learning Approach for Prognostication in Canine Pulmonary Carcinoma to Enhance Reproducibility.

The integration of deep learning-based tools into diagnostic workflows is increasingly prevalent due to their efficiency and reproducibility in various settings. We investigated the utility of automated nuclear morphometry for assessing nuclear pleomorphism (NP), a criterion of malignancy in the current grading system in canine pulmonary carcinoma (cPC), and its prognostic implications. We developed a deep learning-based algorithm for evaluating NP (variation in size, i.e., anisokaryosis and/or shape) using a segmentation model. Its performance was evaluated on 46 cPC cases with comprehensive follow-up data regarding its accuracy in nuclear segmentation and its prognostic ability. Its assessment of NP was compared to manual morphometry and established prognostic tests (pathologists' NP estimates (n = 11), mitotic count, histological grading, and TNM-stage). The standard deviation (SD) of the nuclear area, indicative of anisokaryosis, exhibited good discriminatory ability for tumor-specific survival, with an area under the curve (AUC) of 0.80 and a hazard ratio (HR) of 3.38. The algorithm achieved values comparable to manual morphometry. In contrast, the pathologists' estimates of anisokaryosis resulted in HR values ranging from 0.86 to 34.8, with slight inter-observer reproducibility (k = 0.204). Other conventional tests had no significant prognostic value in our study cohort. Fully automated morphometry promises a time-efficient and reproducible assessment of NP with a high prognostic value. Further refinement of the algorithm, particularly to address undersegmentation, and application to a larger study population are required.

Automated diagnosis of 7 canine skin tumors using machine learning on H&E-stained whole slide images.

Microscopic evaluation of hematoxylin and eosin-stained slides is still the diagnostic gold standard for a variety of diseases, including neoplasms. Nevertheless, intra- and interrater variability are well documented among pathologists. So far, computer assistance via automated image analysis has shown potential to support pathologists in improving accuracy and reproducibility of quantitative tasks. In this proof of principle study, we describe a machine-learning-based algorithm for the automated diagnosis of 7 of the most common canine skin tumors: trichoblastoma, squamous cell carcinoma, peripheral nerve sheath tumor, melanoma, histiocytoma, mast cell tumor, and plasmacytoma. We selected, digitized, and annotated 350 hematoxylin and eosin-stained slides (50 per tumor type) to create a database divided into training, n = 245 whole-slide images (WSIs), validation (n = 35 WSIs), and test sets (n = 70 WSIs). Full annotations included the 7 tumor classes and 6 normal skin structures. The data set was used to train a convolutional neural network (CNN) for the automatic segmentation of tumor and nontumor classes. Subsequently, the detected tumor regions were classified patch-wise into 1 of the 7 tumor classes. A majority of patches-approach led to a tumor classification accuracy of the network on the slide-level of 95% (133/140 WSIs), with a patch-level precision of 85%. The same 140 WSIs were provided to 6 experienced pathologists for diagnosis, who achieved a similar slide-level accuracy of 98% (137/140 correct majority votes). Our results highlight the feasibility of artificial intelligence-based methods as a support tool in diagnostic oncologic pathology with future applications in other species and tumor types.

Computer-assisted mitotic count using a deep learning-based algorithm improves interobserver reproducibility and accuracy.

The mitotic count (MC) is an important histological parameter for prognostication of malignant neoplasms. However, it has inter- and intraobserver discrepancies due to difficulties in selecting the region of interest (MC-ROI) and in identifying or classifying mitotic figures (MFs). Recent progress in the field of artificial intelligence has allowed the development of high-performance algorithms that may improve standardization of the MC. As algorithmic predictions are not flawless, computer-assisted review by pathologists may ensure reliability. In the present study, we compared partial (MC-ROI preselection) and full (additional visualization of MF candidates and display of algorithmic confidence values) computer-assisted MC analysis to the routine (unaided) MC analysis by 23 pathologists for whole-slide images of 50 canine cutaneous mast cell tumors (ccMCTs). Algorithmic predictions aimed to assist pathologists in detecting mitotic hotspot locations, reducing omission of MFs, and improving classification against imposters. The interobserver consistency for the MC significantly increased with computer assistance (interobserver correlation coefficient, ICC = 0.92) compared to the unaided approach (ICC = 0.70). Classification into prognostic stratifications had a higher accuracy with computer assistance. The algorithmically preselected hotspot MC-ROIs had a consistently higher MCs than the manually selected MC-ROIs. Compared to a ground truth (developed with immunohistochemistry for phosphohistone H3), pathologist performance in detecting individual MF was augmented when using computer assistance (F1-score of 0.68 increased to 0.79) with a reduction in false negatives by 38%. The results of this study demonstrate that computer assistance may lead to more reproducible and accurate MCs in ccMCTs.

Deep learning algorithms out-perform veterinary pathologists in detecting the mitotically most active tumor region.

Manual count of mitotic figures, which is determined in the tumor region with the highest mitotic activity, is a key parameter of most tumor grading schemes. It can be, however, strongly dependent on the area selection due to uneven mitotic figure distribution in the tumor section. We aimed to assess the question, how significantly the area selection could impact the mitotic count, which has a known high inter-rater disagreement. On a data set of 32 whole slide images of H&E-stained canine cutaneous mast cell tumor, fully annotated for mitotic figures, we asked eight veterinary pathologists (five board-certified, three in training) to select a field of interest for the mitotic count. To assess the potential difference on the mitotic count, we compared the mitotic count of the selected regions to the overall distribution on the slide. Additionally, we evaluated three deep learning-based methods for the assessment of highest mitotic density: In one approach, the model would directly try to predict the mitotic count for the presented image patches as a regression task. The second method aims at deriving a segmentation mask for mitotic figures, which is then used to obtain a mitotic density. Finally, we evaluated a two-stage object-detection pipeline based on state-of-the-art architectures to identify individual mitotic figures. We found that the predictions by all models were, on average, better than those of the experts. The two-stage object detector performed best and outperformed most of the human pathologists on the majority of tumor cases. The correlation between the predicted and the ground truth mitotic count was also best for this approach (0.963-0.979). Further, we found considerable differences in position selection between pathologists, which could partially explain the high variance that has been reported for the manual mitotic count. To achieve better inter-rater agreement, we propose to use a computer-based area selection for support of the pathologist in the manual mitotic count.

[Detection of BRAF mutation in canine prostatic diseases]. / Nachweis der BRAF-Mutation bei kaninen Prostataerkrankungen.

OBJECTIVE: In the literature, the BRAF mutation is reported to have been identified in 80 % of the examined canine prostate carcinomas (PCa). The objectives of this study were to test for the BRAF mutation in canine PCa in our cohort of canine patients, to determine the specificity and sensitivity of the test for this mutation, as well as to identify the association between the presence of the BRAF mutation and the histologic picture of PCa. Moreover, the method was to be established in cytology samples. MATERIAL AND METHODS: Biopsy samples (n = 70) and cytologic slides (n = 17) of 87 dogs with prostatic diseases were selected. Prostatic diseases were classified according to the literature as benign prostate hyperplasia (BPH, n = 22), prostatitis (n = 14), squamous cell metaplasia of the prostate (PM, n = 2), atrophy following castration (n = 3) und PCa (n = 46; histologic diagnosis n = 35, cytologic diagnosis n = 11). Additionally, the Gleason score was determined for each PCa. DNA isolation was performed using commercially available kits. Exon 15 was examined using the TaqMan® SNP assay. The specificity and sensitivity of the test were calculated. RESULTS: A Gleason score of 6 and 7 was shown in 1 PCa each, in 33 cases the score ranged between 8 and 10. Sufficient amount of good-quality DNA was isolated from all samples. 28/46 PCa were tested positive for the BRAF mutation (sensitivity 61 %). The BRAF mutation was not evident in any of the dogs with BPH, prostatitis, PM or atrophy (specificity 100 %). PCa positive for the BRAF mutation exhibited a significantly higher Gleason score (p = 0.002) in comparison to PCa without this mutation. CONCLUSION AND CLINICAL RELEVANCE: BRAF mutation analysis is a highly specific method and may aid in confirming the diagnosis of PCa in histologically and cytologically questionable cases. PCa positive for BRAF mutation exhibited more criteria of malignancy than PCa without this mutation. The clinical, therapeutic, and prognostic relevance of these findings needs to be evaluated by further studies.

Aging and Senescence in Canine Testes.

Senescent cells accumulate with age but tissue-based studies of senescent cells are limited to selected organs from humans, mice, and primates. Cell culture and xenograft studies have indicated that senescent cells in the microenvironment may play a role in tumor proliferation via paracrine activities. Dogs develop age-related conditions, including in the testis, but cellular senescence has not been confirmed. We hypothesized that senescent cells accumulate with age in canine testes and in the microenvironment of testicular tumors. We tested the expression of the established senescence markers γH2AX and p21 on normal formalin-fixed, paraffin-embedded testes from 15 young dogs (<18 months of age) and 15 old dogs (7-15 years of age) and correlated the findings with age-dependent morphological changes. A statistically significant age-dependent increase in the percentage of p21-expressing cells was observed for testicular fibroblasts (4-fold) and Leydig cells (8-fold). However, p21-expressing cells were still a rare event. In contrast, the percentage of γH2AX-positive cells did not increase with age. P21- and γH2AX-expressing cells were rare in the microenvironments of tumors. Age-dependent morphological changes included an increased mean number of Leydig cells per intertubular triangle (2.95-fold) and a decreased spermatogenesis score. To our surprise, no age-related changes were recorded for interstitial collagen content, mean tubular diameter, and epithelial area. Opposed to our expectations based on previous in vitro data, we did not identify evidence of a correlation between age-associated accumulation of senescent cells and testicular tumor development. Understanding the role of the microenvironment in senescence obviously remains a challenging task.

Correlation of BRAF Variant V595E, Breed, Histological Grade and Cyclooxygenase-2 Expression in Canine Transitional Cell Carcinomas.

The presence of BRAF variant V595E, as well as an increased cyclooxygenase-2 (COX-2) expression in canine transitional cell carcinoma (TCC) are well-described in the literature. The aim of the present study was to investigate the correlation between breed (terrier versus non-terrier dogs), histological grade, COX-2 expression, and BRAF mutation in canine TCC. Therefore, transmural TCC biopsies from 65 dogs (15 terriers, 50 non-terriers) were graded histologically into low- and high-grade. Immunohistochemical evaluation of the intensity of COX-2 expression was performed using an immunoreactive score (IRS). Exon 15 of chromosome 16 was examined for the BRAF variant c.1799T>A by TaqMan® SNP assay. TCC was low-grade in 20 cases (one terrier, 19 non-terriers) and high-grade in 45 cases (14 terriers, 31 non-terriers). Contrary to humans, histological grade was not significantly correlated to the intensity of COX-2 expression. BRAF mutation was detected in 11/15 (73%) TCC of terriers and in 18/50 (36%) TCC of non-terriers. Histological grade and BRAF mutation were not correlated significantly (p = 0.2912). Terriers had a considerably higher prevalence of high-grade tumors (p < 0.0001), as well as of BRAF mutation (p ≤ 0.05) compared to non-terriers. In non-terriers, neoplasms with BRAF mutation showed a significantly higher intensity of COX-2 expression than those without BRAF mutation (p ≤ 0.05). In conclusion, in contrast to humans, testing for BRAF mutation in canine TCC is a sensitive diagnostic method especially in terriers (73%) and may be recommended as a screening test. However, evidence of BRAF mutation in canine TCC is not a predictor for the histological grade. Moreover, a positive correlation between histological grade and the intensity of COX-2 expression was not found. Further studies are necessary to clarify the clinical and prognostic relevance of the elevated intensity of COX-2 expression of TCC with BRAF mutation detected in non-terriers.