Purpose: The study of speckle from imaging systems has a rich history, and recently it was proposed that a fractal or power law distribution of scatterers in vascularized tissue will lead to a form of the Burr probability distribution functions for speckle amplitudes. This hypothesis is generalized and tested in theory, simulations, and experiments.

Approach: We argue that two broadly applicable conjectures are sufficient to justify the applicability of the Burr distribution for speckle from a number of acoustical, optical, and other pulse-echo systems. The first requirement is a multiscale power law distribution of weak scatterers, and the second is a linear approximation for the increase in echo intensity with size over some range of applicability.

Results: The Burr distribution for speckle emerges under a wide variety of conditions and system parameters, and from this one can estimate the governing power law parameter, commonly in the range of 2 to 6. However, system effects including the imaging point spread function and the degree of focusing will influence the Burr parameters.

Conclusions: A generalized pair of conditions is sufficient for producing Burr distributions across a number of imaging systems. Simulations and some theoretical considerations indicate that the estimated Burr power law parameter will increase with increasing density of scatters. For studies of speckle from living tissue or multiscale natural structures, the Burr distribution should be considered as a long tail alternative to classical distributions.

Purpose: An accurate zonal segmentation of the prostate is required for prostate cancer (PCa) management with MRI.

Approach: The aim of this work is to present UFNet, a deep learning-based method for automatic zonal segmentation of the prostate from T2-weighted (T2w) MRI. It takes into account the image anisotropy, includes both spatial and channelwise attention mechanisms and uses loss functions to enforce prostate partition. The method was applied on a private multicentric three-dimensional T2w MRI dataset and on the public two-dimensional T2w MRI dataset ProstateX. To assess the model performance, the structures segmented by the algorithm on the private dataset were compared with those obtained by seven radiologists of various experience levels.

Results: On the private dataset, we obtained a Dice score (DSC) of 93.90  ±  2.85 for the whole gland (WG), 91.00  ±  4.34 for the transition zone (TZ), and 79.08  ±  7.08 for the peripheral zone (PZ). Results were significantly better than other compared networks’ (p-value  <  0.05). On ProstateX, we obtained a DSC of 90.90  ±  2.94 for WG, 86.84  ±  4.33 for TZ, and 78.40  ±  7.31 for PZ. These results are similar to state-of-the art results and, on the private dataset, are coherent with those obtained by radiologists. Zonal locations and sectorial positions of lesions annotated by radiologists were also preserved.

Conclusions: Deep learning-based methods can provide an accurate zonal segmentation of the prostate leading to a consistent zonal location and sectorial position of lesions, and therefore can be used as a helping tool for PCa diagnosis.

Purpose: Accurate segmentation of the pancreas using abdominal computed tomography (CT) scans is a prerequisite for a computer-aided diagnosis system to detect pathologies and perform quantitative assessment of pancreatic disorders. Manual outlining of the pancreas is tedious, time-consuming, and prone to subjective errors, and thus clearly not a viable solution for large datasets.

Approach: We introduce a multiphase morphology-guided deep learning framework for efficient three-dimensional segmentation of the pancreas in CT images. The methodology works by localizing the pancreas using a modified visual geometry group-19 architecture, which is a 19-layer convolutional neural network model that helped reduce the region of interest for more efficient computation and removed most of the peripheral structures from consideration during the segmentation process. Subsequently, soft labels for segmentation of the pancreas in the localized region were generated using the U-net model. Finally, the model integrates the morphology prior of the pancreas to update soft labels and perform segmentation. The morphology prior is a single three-dimensional matrix, defined over the general shape and size of the pancreases from multiple CT abdominal images, that helps improve segmentation of the pancreas.

Results: The system was trained and tested on the National Institutes of Health dataset (82 CT scans of the healthy pancreas). In fourfold cross-validation, the system produced an average Dice–SØrensen coefficient of 88.53% and outperformed state-of-the-art techniques.

Conclusions: Localizing the pancreas assists in reducing segmentation errors and eliminating peripheral structures from consideration. Additionally, the morphology-guided model efficiently improves the overall segmentation of the pancreas.

Purpose: We propose a super-resolution (SR) method, named SR-CycleGAN, for SR of clinical computed tomography (CT) images to the micro-focus x-ray CT CT (μCT) level. Due to the resolution limitations of clinical CT (about 500  ×  500  ×  500  μm³  /  voxel), it is challenging to obtain enough pathological information. On the other hand, μCT scanning allows the imaging of lung specimens with significantly higher resolution (about 50  ×  50  ×  50  μm³  /  voxel or higher), which allows us to obtain and analyze detailed anatomical information. As a way to obtain detailed information such as cancer invasion and bronchioles from preoperative clinical CT images of lung cancer patients, the SR of clinical CT images to the μCT level is desired.

Approach: Typical SR methods require aligned pairs of low-resolution (LR) and high-resolution images for training, but it is infeasible to obtain precisely aligned paired clinical CT and μCT images. To solve this problem, we propose an unpaired SR approach that can perform SR on clinical CT to the μCT level. We modify a conventional image-to-image translation network named CycleGAN to an inter-modality translation network named SR-CycleGAN. The modifications consist of three parts: (1) an innovative loss function named multi-modality super-resolution loss, (2) optimized SR network structures for enlarging the input LR image to k²-times by width and height to obtain the SR output, and (3) sub-pixel shuffling layers for reducing computing time.

Results: Experimental results demonstrated that our method successfully performed SR of lung clinical CT images. SSIM and PSNR scores of our method were 0.54 and 17.71, higher than the conventional CycleGAN’s scores of 0.05 and 13.64, respectively.

Conclusions: The proposed SR-CycleGAN is usable for the SR of a lung clinical CT into μCT scale, while conventional CycleGAN output images with low qualitative and quantitative values. More lung micro-anatomy information could be observed to aid diagnosis, such as the shape of bronchioles walls.

Purpose: Nephrologists have empirically predicted renal function from renal morphology. In diagnosing a case of renal dysfunction of unknown course, acute kidney injury and chronic kidney disease are diagnosed from blood tests and an imaging study including magnetic resonance imaging (MRI), and an examination/treatment policy is determined. A framework for the estimation of renal function from water images obtained using the Dixon method is proposed to provide information that helps clinicians reach a diagnosis by accurately estimating renal function on the basis of renal MRI.

Approach: The proposed framework consists of four steps. First, the kidney area is extracted by MRI using the Dixon method with a U-net by deep learning. Second, the extracted renal region is registered with the target mask. Third, the kidney features are calculated based on the target mask classification information created by a specialist. Fourth, the estimated glomerular filtration rate (eGFR) representing the renal function is estimated using a regression support vector machine from the calculated features.

Results: For the accuracy evaluation, we conducted an experiment to estimate the eGFR when MRI was performed and the eGFR slope, which is the annual rate of decline in eGFR. When the accuracy was evaluated for 165 subjects, the eGFR was estimated to have a root mean square error (RMSE) of 11.99 and a correlation coefficient of 0.83. Moreover, the eGFR slope was estimated to have an RMSE of 4.8 and a correlation coefficient of 0.5.

Conclusions: Therefore, the proposed method shows the possibility of estimating the prognosis of renal function based on water images obtained by the Dixon method.

Purpose: Implanting stents to re-open stenotic lesions during percutaneous coronary interventions is considered a standard treatment for acute or chronic coronary syndrome. Intravascular ultrasound (IVUS) can be used to guide and assess the technical success of these interventions. Automatically segmenting stent struts in IVUS sequences improves workflow efficiency but is non-trivial due to a challenging image appearance entailing manifold ambiguities with other structures. Manual, ungated IVUS pullbacks constitute a challenge in this context. We propose a fully data-driven strategy to first longitudinally detect and subsequently segment stent struts in IVUS frames.

Approach: A cascaded deep learning approach is presented. It first trains an encoder model to classify frames as “stent,” “no stent,” or “no use.” A segmentation model then delineates stent struts on a pixel level only in frames with a stent label. The first stage of the cascade acts as a gateway to reduce the risk for false positives in the second stage, the segmentation, which is trained on a smaller and difficult-to-annotate dataset. Training of the classification and segmentation model was based on 49,888 and 1826 frames of 74 sequences from 35 patients, respectively.

Results: The longitudinal classification yielded Dice scores of 92.96%, 82.35%, and 94.03% for the classes stent, no stent, and no use, respectively. The segmentation achieved a Dice score of 65.1% on the stent ground truth (intra-observer performance: 75.5%) and 43.5% on all frames (including frames without stent, with guidewires, calcium, or without clinical use). The latter improved to 49.5% when gating the frames by the classification decision and further increased to 57.4% with a heuristic on the plausible stent strut area.

Conclusions: A data-driven strategy for segmenting stents in ungated, manual pullbacks was presented—the most common and practical scenario in the time-critical clinical workflow. We demonstrated a mitigated risk for ambiguities and false positive predictions.

Purpose: An open question in deep clustering is how to explain what in the image is driving the cluster assignments. This is especially important for applications in medical imaging when the derived cluster assignments may inform decision-making or create new disease subtypes. We develop cluster activation mapping (CLAM), which is methodology to create localization maps highlighting the image regions important for cluster assignment.

Approach: Our approach uses a linear combination of the activation channels from the last layer of the encoder within a pretrained autoencoder. The activation channels are weighted by a channelwise confidence measure, which is a modification of score-CAM.

Results: Our approach performs well under medical imaging-based simulation experiments, when the image clusters differ based on size, location, and intensity of abnormalities. Under simulation, the cluster assignments were predicted with 100% accuracy when the number of clusters was set at the true value. In addition, applied to computed tomography scans from a sarcoidosis population, CLAM identified two subtypes of sarcoidosis based purely on CT scan presentation, which were significantly associated with pulmonary function tests and visual assessment scores, such as ground-glass, fibrosis, and honeycombing.

Conclusions: CLAM is a transparent methodology for identifying explainable groupings of medical imaging data. As deep learning networks are often criticized and not trusted due to their lack of interpretability, our contribution of CLAM to deep clustering architectures is critical to our understanding of cluster assignments, which can ultimately lead to new subtypes of diseases.

Purpose: Automatic outlining of different tissue types in digitized histological specimen provides a basis for follow-up analyses and can potentially guide subsequent medical decisions. The immense size of whole-slide-images (WSIs), however, poses a challenge in terms of computation time. In this regard, the analysis of nonoverlapping patches outperforms pixelwise segmentation approaches but still leaves room for optimization. Furthermore, the division into patches, regardless of the biological structures they contain, is a drawback due to the loss of local dependencies.

Approach: We propose to subdivide the WSI into coherent regions prior to classification by grouping visually similar adjacent pixels into superpixels. Afterward, only a random subset of patches per superpixel is classified and patch labels are combined into a superpixel label. We propose a metric for identifying superpixels with an uncertain classification and evaluate two medical applications, namely tumor area and invasive margin estimation and tumor composition analysis.

Results: The algorithm has been developed on 159 hand-annotated WSIs of colon resections and its performance is compared with an analysis without prior segmentation. The algorithm shows an average speed-up of 41% and an increase in accuracy from 93.8% to 95.7%. By assigning a rejection label to uncertain superpixels, we further increase the accuracy by 0.4%. While tumor area estimation shows high concordance to the annotated area, the analysis of tumor composition highlights limitations of our approach.

Conclusion: By combining superpixel segmentation and patch classification, we designed a fast and accurate framework for whole-slide cartography that is AI-model agnostic and provides the basis for various medical endpoints.