Radiation dose reduction of 50% in dynamic myocardial CT perfusion with skipped beat acquisition: a retrospective study.

BACKGROUND: Dynamic myocardial computed tomography perfusion (CTP) is a novel imaging technique that increases the applicability of CT for cardiac imaging; however, the scanning requires a substantial radiation dose. PURPOSE: To investigate the feasibility of dose reduction in dynamic CTP by comparing all-heartbeat acquisitions to periodic skipping of heartbeats. MATERIAL AND METHODS: We retrieved imaging data of 38 dynamic CTP patients and created new datasets with every fourth, third or second beat (Skip1:4, Skip1:3, Skip1:2, respectively) removed. Seven observers evaluated the resulting images and perfusion maps for perfusion deficits. The mean blood flow (MBF) in each of the 16 myocardial segments was compared per skipped-beat level, normalized by the respective MBF for the full dose, and averaged across patients. The number of segments/cases whose MBF was <1.0 mL/g/min were counted. RESULTS: Out of 608 segments in 38 cases, the total additional number of false-negative (FN) segments over those present in the full-dose acquisitions and the number of additional false-positive cases were shown as acquisition (segment [%], case): Skip1:4: 7 (1.2%, 1); Skip1:3: 12 (2%, 3), and Skip1:2: 5 (0.8%, 2). The variability in quantitative MBF analysis in the repeated analysis for the reference condition resulted in 8 (1.3%) additional FN segments. The normalized results show a comparable MBF across all segments and patients, with relative mean MBFs as 1.02 ± 0.16, 1.03 ± 0.25, and 1.06 ± 0.30 for the Skip1:4, Skip1:3, and Skip1:2 protocols, respectively. CONCLUSION: Skipping every second beat acquisition during dynamic myocardial CTP appears feasible and may result in a radiation dose reduction of 50%. Diagnostic performance does not decrease after removing 50% of time points in dynamic sequence.

Improving classification accuracy of fine-tuned CNN models: Impact of hyperparameter optimization.

The immense popularity of convolutional neural network (CNN) models has sparked a growing interest in optimizing their hyperparameters. Discovering the ideal values for hyperparameters to achieve optimal CNN training is a complex and time-consuming task, often requiring repetitive numerical experiments. As a result, significant attention is currently being devoted to developing methods aimed at tailoring hyperparameters for specific CNN models and classification tasks. While existing optimization methods often yield favorable image classification results, they do not provide guidance on which hyperparameters are worth optimizing, the appropriate value ranges for those hyperparameters, or whether it is reasonable to use a subset of training data for the optimization process. This work is focused on the optimization of hyperparameters during transfer learning, with the goal of investigating how different optimization methods and hyperparameter selections impact the performance of fine-tuned models. In our experiments, we assessed the importance of various hyperparameters and identified the ranges within which optimal CNN training can be achieved. Additionally, we compared four hyperparameter optimization methods-grid search, random search, Bayesian optimization, and the Asynchronous Successive Halving Algorithm (ASHA). We also explored the feasibility of fine-tuning hyperparameters using a subset of the training data. By optimizing the hyperparameters, we observed an improvement in CNN classification accuracy of up to 6%. Furthermore, we found that achieving a balance in class distribution within the subset of data used for parameter optimization is crucial in establishing the optimal set of hyperparameters for CNN training. The results we obtained demonstrate that hyperparameter optimization is highly dependent on the specific task and dataset at hand.

Localization of spleen and kidney organs from CT scans based on classification of slices in rotational views.

This article presents a novel multiple organ localization and tracking technique applied to spleen and kidney regions in computed tomography images. The proposed solution is based on a unique approach to classify regions in different spatial projections (e.g., side projection) using convolutional neural networks. Our procedure merges classification results from different projection resulting in a 3D segmentation. The proposed system is able to recognize the contour of the organ with an accuracy of 88-89% depending on the body organ. Research has shown that the use of a single method can be useful for the detection of different organs: kidney and spleen. Our solution can compete with U-Net based solutions in terms of hardware requirements, as it has significantly lower demands. Additionally, it gives better results in small data sets. Another advantage of our solution is a significantly lower training time on an equally sized data set and more capabilities to parallelize calculations. The proposed system enables visualization, localization and tracking of organs and is therefore a valuable tool in medical diagnostic problems.

Multireader image quality evaluation of dynamic myocardial computed tomography perfusion imaging with a novel four-dimensional noise reduction filter.

BACKGROUND: Dynamic myocardial computed tomography perfusion (CTP) is a novel technique able to depict cardiac ischemia. PURPOSE: To evaluate the impact of a four-dimensional noise reduction filter (similarity filter [4D-SF]) on image quality in dynamic CTP imaging, allowing for substantial radiation dose reduction. MATERIAL AND METHODS: Dynamic CTP datasets of 30 patients (16 women) with suspected coronary artery disease, acquired with a 320-slice CT system, were retrieved, reconstructed with the deep learning-based algorithm of the system (DLR), and filtered with the 4D-SF. For each case, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in six regions of interest (33-38mm2) were calculated before and after filtering, in four-chamber and short-axis views, and t-tested. Furthermore, six radiologists of different expertise evaluated subjective image preference by answering five visual grading analysis-type questions (regarding acceptable level of noise, absence of artifacts, natural appearance, cardiac contour sharpness, diagnostic acceptability) using a 5-point scale. The results were analyzed using visual grade characteristics (VGC) and intraclass correlation coefficient (ICC). RESULTS: Mean SNR in four-chamber view (unfiltered vs. filtered) were: septum=4.1 ± 2.1 versus 7.6 ± 5.6; lateral wall=4.5 ± 2.0 versus 8.0 ± 4.9; CNRseptum=16.6 ± 8.9 versus 31.7 ± 28; lateral wall=16.2 ± 8.9 versus 31.3 ± 28.9. Similar results were obtained in short-axis view. The perceived filtered image quality indicated decreased noise (VGCAUC=0.96) and artifacts (0.65), improved natural appearance (0.59), cardiac contour sharpness (0.74), and diagnostic acceptability (0.78). The inter-observer variability was excellent (ICC=0.79). All results were statistically significant (P < 0.05). CONCLUSION: Similarity filtering after DLR improves image quality, possibly enabling dose reduction in dynamic CTP imaging in patient with suspected chronic coronary syndrome.

Convolutional Neural Networks for the Evaluation of Chronic and Inflammatory Lesions in Kidney Transplant Biopsies.

In kidney transplant biopsies, both inflammation and chronic changes are important features that predict long-term graft survival. Quantitative scoring of these features is important for transplant diagnostics and kidney research. However, visual scoring is poorly reproducible and labor intensive. The goal of this study was to investigate the potential of convolutional neural networks (CNNs) to quantify inflammation and chronic features in kidney transplant biopsies. A structure segmentation CNN and a lymphocyte detection CNN were applied on 125 whole-slide image pairs of periodic acid-Schiff- and CD3-stained slides. The CNN results were used to quantify healthy and sclerotic glomeruli, interstitial fibrosis, tubular atrophy, and inflammation within both nonatrophic and atrophic tubuli, and in areas of interstitial fibrosis. The computed tissue features showed high correlation with Banff lesion scores of five pathologists (A.A., A.Dend., J.H.B., J.K., and T.N.). Analyses on a small subset showed a moderate correlation toward higher CD3+ cell density within scarred regions and higher CD3+ cell count inside atrophic tubuli correlated with long-term change of estimated glomerular filtration rate. The presented CNNs are valid tools to yield objective quantitative information on glomeruli number, fibrotic tissue, and inflammation within scarred and non-scarred kidney parenchyma in a reproducible manner. CNNs have the potential to improve kidney transplant diagnostics and will benefit the community as a novel method to generate surrogate end points for large-scale clinical studies.

Quantitative assessment of inflammatory infiltrates in kidney transplant biopsies using multiplex tyramide signal amplification and deep learning.

Delayed graft function (DGF) is a strong risk factor for development of interstitial fibrosis and tubular atrophy (IFTA) in kidney transplants. Quantitative assessment of inflammatory infiltrates in kidney biopsies of DGF patients can reveal predictive markers for IFTA development. In this study, we combined multiplex tyramide signal amplification (mTSA) and convolutional neural networks (CNNs) to assess the inflammatory microenvironment in kidney biopsies of DGF patients (n = 22) taken at 6 weeks post-transplantation. Patients were stratified for IFTA development (<10% versus ≥10%) from 6 weeks to 6 months post-transplantation, based on histopathological assessment by three kidney pathologists. One mTSA panel was developed for visualization of capillaries, T- and B-lymphocytes and macrophages and a second mTSA panel for T-helper cell and macrophage subsets. The slides were multi spectrally imaged and custom-made python scripts enabled conversion to artificial brightfield whole-slide images (WSI). We used an existing CNN for the detection of lymphocytes with cytoplasmatic staining patterns in immunohistochemistry and developed two new CNNs for the detection of macrophages and nuclear-stained lymphocytes. F1-scores were 0.77 (nuclear-stained lymphocytes), 0.81 (cytoplasmatic-stained lymphocytes), and 0.82 (macrophages) on a test set of artificial brightfield WSI. The CNNs were used to detect inflammatory cells, after which we assessed the peritubular capillary extent, cell density, cell ratios, and cell distance in the two patient groups. In this cohort, distance of macrophages to other immune cells and peritubular capillary extent did not vary significantly at 6 weeks post-transplantation between patient groups. CD163+ cell density was higher in patients with ≥10% IFTA development 6 months post-transplantation (p < 0.05). CD3+CD8-/CD3+CD8+ ratios were higher in patients with <10% IFTA development (p < 0.05). We observed a high correlation between CD163+ and CD4+GATA3+ cell density (R = 0.74, p < 0.001). Our study demonstrates that CNNs can be used to leverage reliable, quantitative results from mTSA-stained, multi spectrally imaged slides of kidney transplant biopsies.

Optimized tumour infiltrating lymphocyte assessment for triple negative breast cancer prognostics.

The tumour microenvironment has been shown to be a valuable source of prognostic information for different cancer types. This holds in particular for triple negative breast cancer (TNBC), a breast cancer subtype for which currently no prognostic biomarkers are established. Although different methods to assess tumour infiltrating lymphocytes (TILs) have been published, it remains unclear which method (marker, region) yields the most optimal prognostic information. In addition, to date, no objective TILs assessment methods are available. For this proof of concept study, a subset of our previously described TNBC cohort (n = 94) was stained for CD3, CD8 and FOXP3 using multiplex immunohistochemistry and subsequently imaged by a multispectral imaging system. Advanced whole-slide image analysis algorithms, including convolutional neural networks (CNN) were used to register unmixed multispectral images and corresponding H&E sections, to segment the different tissue compartments (tumour, stroma) and to detect all individual positive lymphocytes. Densities of positive lymphocytes were analysed in different regions within the tumour and its neighbouring environment and correlated to relapse free survival (RFS) and overall survival (OS). We found that for all TILs markers the presence of a high density of positive cells correlated with an improved survival. None of the TILs markers was superior to the others. The results of TILs assessment in the various regions did not show marked differences between each other. The negative correlation between TILs and survival in our cohort are in line with previous studies. Our results provide directions for optimizing TILs assessment methodology.

A U-Net based framework to quantify glomerulosclerosis in digitized PAS and H&E stained human tissues.

Reliable counting of glomeruli and evaluation of glomerulosclerosis in renal specimens are essential steps to assess morphological changes in kidney and identify individuals requiring treatment. Because microscopic identification of sclerosed glomeruli performed under the microscope is labor intensive, we developed a deep learning (DL) approach to identify and classify glomeruli as normal or sclerosed in digital whole slide images (WSIs). The segmentation and classification of glomeruli was performed by the U-Net model. Subsequently, glomerular classifications were refined based on glomerular histomorphometry. The U-Net model was trained using patches from Periodic Acid-Schiff (PAS) stained WSIs (n=31) from the AIDPATH - a multi-center dataset, and then tested on an independent set of WSIs (n=20) including PAS (n=6), and hematoxylin and eosin (H&E) stained WSIs (n=14) from four other institutions. The training and test WSIs were obtained from formalin fixed and paraffin embedded blocks with of human kidney specimens each presenting various proportions of normal and sclerosed glomeruli. In the PAS stained WSIs, normal and sclerosed glomeruli were respectively classified with the F1-score of 97.5% and 68.8%. In the H&E stained WSIs, the F1-scores of 90.8% and 78.1% were achieved. Regardless the tissue staining, the glomeruli in the test WSIs were classified with the F1-score of 94.5% (n=923, normal) and 76.8% for (n=261, sclerosed). These results demonstrate for the first time that a framework based on the U-Net model trained with glomerular patches from PAS stained WSIs can reliably segment and classify normal and sclerosed glomeruli in PAS and also H&E stained WSIs. Our approach yielded higher accuracy of glomerular classifications than some of the recently published methods. Additionally, our test set of images with ground truth is publicly available.

Artificial intelligence to detect MYC translocation in slides of diffuse large B-cell lymphoma.

In patients with suspected lymphoma, the tissue biopsy provides lymphoma confirmation, classification, and prognostic factors, including genetic changes. We developed a deep learning algorithm to detect MYC rearrangement in scanned histological slides of diffuse large B-cell lymphoma. The H&E-stained slides of 287 cases from 11 hospitals were used for training and evaluation. The overall sensitivity to detect MYC rearrangement was 0.93 and the specificity 0.52, showing that prediction of MYC translocation based on morphology alone was possible in 93% of MYC-rearranged cases. This would allow a simple and fast prescreening, saving approximately 34% of genetic tests with the current algorithm.

Impact of rescanning and normalization on convolutional neural network performance in multi-center, whole-slide classification of prostate cancer.

Algorithms can improve the objectivity and efficiency of histopathologic slide analysis. In this paper, we investigated the impact of scanning systems (scanners) and cycle-GAN-based normalization on algorithm performance, by comparing different deep learning models to automatically detect prostate cancer in whole-slide images. Specifically, we compare U-Net, DenseNet and EfficientNet. Models were developed on a multi-center cohort with 582 WSIs and subsequently evaluated on two independent test sets including 85 and 50 WSIs, respectively, to show the robustness of the proposed method to differing staining protocols and scanner types. We also investigated the application of normalization as a pre-processing step by two techniques, the whole-slide image color standardizer (WSICS) algorithm, and a cycle-GAN based method. For the two independent datasets we obtained an AUC of 0.92 and 0.83 respectively. After rescanning the AUC improves to 0.91/0.88 and after style normalization to 0.98/0.97. In the future our algorithm could be used to automatically pre-screen prostate biopsies to alleviate the workload of pathologists.

Learning to detect lymphocytes in immunohistochemistry with deep learning.

The immune system is of critical importance in the development of cancer. The evasion of destruction by the immune system is one of the emerging hallmarks of cancer. We have built a dataset of 171,166 manually annotated CD3+ and CD8+ cells, which we used to train deep learning algorithms for automatic detection of lymphocytes in histopathology images to better quantify immune response. Moreover, we investigate the effectiveness of four deep learning based methods when different subcompartments of the whole-slide image are considered: normal tissue areas, areas with immune cell clusters, and areas containing artifacts. We have compared the proposed methods in breast, colon and prostate cancer tissue slides collected from nine different medical centers. Finally, we report the results of an observer study on lymphocyte quantification, which involved four pathologists from different medical centers, and compare their performance with the automatic detection. The results give insights on the applicability of the proposed methods for clinical use. U-Net obtained the highest performance with an F1-score of 0.78 and the highest agreement with manual evaluation (κ=0.72), whereas the average pathologists agreement with reference standard was κ=0.64. The test set and the automatic evaluation procedure are publicly available at lyon19.grand-challenge.org.

Convolutional neural networks can accurately distinguish four histologic growth patterns of lung adenocarcinoma in digital slides.

During the diagnostic workup of lung adenocarcinomas (LAC), pathologists evaluate distinct histological tumor growth patterns. The percentage of each pattern on multiple slides bears prognostic significance. To assist with the quantification of growth patterns, we constructed a pipeline equipped with a convolutional neural network (CNN) and soft-voting as the decision function to recognize solid, micropapillary, acinar, and cribriform growth patterns, and non-tumor areas. Slides of primary LAC were obtained from Cedars-Sinai Medical Center (CSMC), the Military Institute of Medicine in Warsaw and the TCGA portal. Several CNN models trained with 19,924 image tiles extracted from 78 slides (MIMW and CSMC) were evaluated on 128 test slides from the three sites by F1-score and accuracy using manual tumor annotations by pathologist. The best CNN yielded F1-scores of 0.91 (solid), 0.76 (micropapillary), 0.74 (acinar), 0.6 (cribriform), and 0.96 (non-tumor) respectively. The overall accuracy of distinguishing the five tissue classes was 89.24%. Slide-based accuracy in the CSMC set (88.5%) was significantly better (p < 2.3E-4) than the accuracy in the MIMW (84.2%) and TCGA (84%) sets due to superior slide quality. Our model can work side-by-side with a pathologist to accurately quantify the percentages of growth patterns in tumors with mixed LAC patterns.

Image processing methods for the structural detection and gradation of placental villi.

The context-based examination of stained tissue specimens is one of the most important procedures in histopathological practice. The development of image processing methods allows for the automation of this process. We propose a method of automatic segmentation of placental structures and assessment of edema present in placental structures from a spontaneous miscarriage. The presented method is based on texture analysis, mathematical morphology, and region growing operations that are applicable to the heterogeneous microscopic images representing histological slides of the placenta. The results presented in this study were obtained using a set of 50 images of single villi originating from 13 histological slides and was compared with the manual evaluation of the pathologist. In the presented experiments, various structures, such as villi, villous mesenchyme, trophoblast, collagen, and vessels have been recognized. Moreover, the gradation of villous edema for three classes (no villous edema, moderate villous edema, and massive villous edema) has been conducted. Villi images were correctly identified in 98.21%, villous mesenchyme was correctly identified in 83.95%, and the villi evaluation was correct in 74% for the edema degree and 86% for the number of vessels. The presented segmentation method may serve as a support for current manual diagnosis methods and reduce the bias related to individual, subjective assessment of experts.

Content-based analysis of Ki-67 stained meningioma specimens for automatic hot-spot selection.

BACKGROUND: Hot-spot based examination of immunohistochemically stained histological specimens is one of the most important procedures in pathomorphological practice. The development of image acquisition equipment and computational units allows for the automation of this process. Moreover, a lot of possible technical problems occur in everyday histological material, which increases the complexity of the problem. Thus, a full context-based analysis of histological specimens is also needed in the quantification of immunohistochemically stained specimens. One of the most important reactions is the Ki-67 proliferation marker in meningiomas, the most frequent intracranial tumour. The aim of our study is to propose a context-based analysis of Ki-67 stained specimens of meningiomas for automatic selection of hot-spots. METHODS: The proposed solution is based on textural analysis, mathematical morphology, feature ranking and classification, as well as on the proposed hot-spot gradual extinction algorithm to allow for the proper detection of a set of hot-spot fields. The designed whole slide image processing scheme eliminates such artifacts as hemorrhages, folds or stained vessels from the region of interest. To validate automatic results, a set of 104 meningioma specimens were selected and twenty hot-spots inside them were identified independently by two experts. The Spearman rho correlation coefficient was used to compare the results which were also analyzed with the help of a Bland-Altman plot. RESULTS: The results show that most of the cases (84) were automatically examined properly with two fields of view with a technical problem at the very most. Next, 13 had three such fields, and only seven specimens did not meet the requirement for the automatic examination. Generally, the Automatic System identifies hot-spot areas, especially their maximum points, better. Analysis of the results confirms the very high concordance between an automatic Ki-67 examination and the expert's results, with a Spearman rho higher than 0.95. CONCLUSION: The proposed hot-spot selection algorithm with an extended context-based analysis of whole slide images and hot-spot gradual extinction algorithm provides an efficient tool for simulation of a manual examination. The presented results have confirmed that the automatic examination of Ki-67 in meningiomas could be introduced in the near future.