Image-level detection of arterial occlusions in 4D-CTA of acute stroke patients using deep learning.

The triage of acute stroke patients is increasingly dependent on four-dimensional CTA (4D-CTA) imaging. In this work, we present a convolutional neural network (CNN) for image-level detection of intracranial anterior circulation artery occlusions in 4D-CTA. The method uses a normalized 3D time-to-signal (TTS) representation of the input image, which is sensitive to differences in the global arrival times caused by the potential presence of vascular pathologies. The TTS map presents the time within the cranial cavity at which the signal reaches a percentage of the maximum signal intensity, corrected for the baseline intensity. The method was trained and validated on (n=214) patient images and tested on an independent set of (n=279) patient images. This test set included all consecutive suspected-stroke patients admitted to our hospital in 2018. The accuracy, sensitivity, and specificity were 92%, 95%, and 92%. The area under the receiver operating characteristics curve was 0.98 (95% CI: 0.95- 0.99). These results show the feasibility of automated stroke triage in 4D-CTA.

Intracerebral Haemorrhage Segmentation in Non-Contrast CT.

A 3-dimensional (3D) convolutional neural network is presented for the segmentation and quantification of spontaneous intracerebral haemorrhage (ICH) in non-contrast computed tomography (NCCT). The method utilises a combination of contextual information on multiple scales for fast and fully automatic dense predictions. To handle a large class imbalance present in the data, a weight map is introduced during training. The method was evaluated on two datasets of 25 and 50 patients respectively. The reference standard consisted of manual annotations for each ICH in the dataset. Quantitative analysis showed a median Dice similarity coefficient of 0.91 [0.87-0.94] and 0.90 [0.85-0.92] for the two test datasets in comparison to the reference standards. Evaluation of a separate dataset of 5 patients for the assessment of the observer variability produced a mean Dice similarity coefficient of 0.95 ± 0.02 for the inter-observer variability and 0.97 ± 0.01 for the intra-observer variability. The average prediction time for an entire volume was 104 ± 15 seconds. The results demonstrate that the method is accurate and approaches the performance of expert manual annotation.

Epithelium segmentation using deep learning in H&E-stained prostate specimens with immunohistochemistry as reference standard.

Given the importance of gland morphology in grading prostate cancer (PCa), automatically differentiating between epithelium and other tissues is an important prerequisite for the development of automated methods for detecting PCa. We propose a new deep learning method to segment epithelial tissue in digitised hematoxylin and eosin (H&E) stained prostatectomy slides using immunohistochemistry (IHC) as reference standard. We used IHC to create a precise and objective ground truth compared to manual outlining on H&E slides, especially in areas with high-grade PCa. 102 tissue sections were stained with H&E and subsequently restained with P63 and CK8/18 IHC markers to highlight epithelial structures. Afterwards each pair was co-registered. First, we trained a U-Net to segment epithelial structures in IHC using a subset of the IHC slides that were preprocessed with color deconvolution. Second, this network was applied to the remaining slides to create the reference standard used to train a second U-Net on H&E. Our system accurately segmented both intact glands and individual tumour epithelial cells. The generalisation capacity of our system is shown using an independent external dataset from a different centre. We envision this segmentation as the first part of a fully automated prostate cancer grading pipeline.

Automated measurement of fetal head circumference using 2D ultrasound images.

In this paper we present a computer aided detection (CAD) system for automated measurement of the fetal head circumference (HC) in 2D ultrasound images for all trimesters of the pregnancy. The HC can be used to estimate the gestational age and monitor growth of the fetus. Automated HC assessment could be valuable in developing countries, where there is a severe shortage of trained sonographers. The CAD system consists of two steps: First, Haar-like features were computed from the ultrasound images to train a random forest classifier to locate the fetal skull. Secondly, the HC was extracted using Hough transform, dynamic programming and an ellipse fit. The CAD system was trained on 999 images and validated on an independent test set of 335 images from all trimesters. The test set was manually annotated by an experienced sonographer and a medical researcher. The reference gestational age (GA) was estimated using the crown-rump length measurement (CRL). The mean difference between the reference GA and the GA estimated by the experienced sonographer was 0.8 ± 2.6, -0.0 ± 4.6 and 1.9 ± 11.0 days for the first, second and third trimester, respectively. The mean difference between the reference GA and the GA estimated by the medical researcher was 1.6 ± 2.7, 2.0 ± 4.8 and 3.9 ± 13.7 days. The mean difference between the reference GA and the GA estimated by the CAD system was 0.6 ± 4.3, 0.4 ± 4.7 and 2.5 ± 12.4 days. The results show that the CAD system performs comparable to an experienced sonographer. The presented system shows similar or superior results compared to systems published in literature. This is the first automated system for HC assessment evaluated on a large test set which contained data of all trimesters of the pregnancy.

Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: The LUNA16 challenge.

Automatic detection of pulmonary nodules in thoracic computed tomography (CT) scans has been an active area of research for the last two decades. However, there have only been few studies that provide a comparative performance evaluation of different systems on a common database. We have therefore set up the LUNA16 challenge, an objective evaluation framework for automatic nodule detection algorithms using the largest publicly available reference database of chest CT scans, the LIDC-IDRI data set. In LUNA16, participants develop their algorithm and upload their predictions on 888 CT scans in one of the two tracks: 1) the complete nodule detection track where a complete CAD system should be developed, or 2) the false positive reduction track where a provided set of nodule candidates should be classified. This paper describes the setup of LUNA16 and presents the results of the challenge so far. Moreover, the impact of combining individual systems on the detection performance was also investigated. It was observed that the leading solutions employed convolutional networks and used the provided set of nodule candidates. The combination of these solutions achieved an excellent sensitivity of over 95% at fewer than 1.0 false positives per scan. This highlights the potential of combining algorithms to improve the detection performance. Our observer study with four expert readers has shown that the best system detects nodules that were missed by expert readers who originally annotated the LIDC-IDRI data. We released this set of additional nodules for further development of CAD systems.

Deep multi-scale location-aware 3D convolutional neural networks for automated detection of lacunes of presumed vascular origin.

Lacunes of presumed vascular origin (lacunes) are associated with an increased risk of stroke, gait impairment, and dementia and are a primary imaging feature of the small vessel disease. Quantification of lacunes may be of great importance to elucidate the mechanisms behind neuro-degenerative disorders and is recommended as part of study standards for small vessel disease research. However, due to the different appearance of lacunes in various brain regions and the existence of other similar-looking structures, such as perivascular spaces, manual annotation is a difficult, elaborative and subjective task, which can potentially be greatly improved by reliable and consistent computer-aided detection (CAD) routines. In this paper, we propose an automated two-stage method using deep convolutional neural networks (CNN). We show that this method has good performance and can considerably benefit readers. We first use a fully convolutional neural network to detect initial candidates. In the second step, we employ a 3D CNN as a false positive reduction tool. As the location information is important to the analysis of candidate structures, we further equip the network with contextual information using multi-scale analysis and integration of explicit location features. We trained, validated and tested our networks on a large dataset of 1075 cases obtained from two different studies. Subsequently, we conducted an observer study with four trained observers and compared our method with them using a free-response operating characteristic analysis. Shown on a test set of 111 cases, the resulting CAD system exhibits performance similar to the trained human observers and achieves a sensitivity of 0.974 with 0.13 false positives per slice. A feasibility study also showed that a trained human observer would considerably benefit once aided by the CAD system.

Improving airway segmentation in computed tomography using leak detection with convolutional networks.

We propose a novel method to improve airway segmentation in thoracic computed tomography (CT) by detecting and removing leaks. Leak detection is formulated as a classification problem, in which a convolutional network (ConvNet) is trained in a supervised fashion to perform the classification task. In order to increase the segmented airway tree length, we take advantage of the fact that multiple segmentations can be extracted from a given airway segmentation algorithm by varying the parameters that influence the tree length and the amount of leaks. We propose a strategy in which the combination of these segmentations after removing leaks can increase the airway tree length while limiting the amount of leaks. This strategy therefore largely circumvents the need for parameter fine-tuning of a given airway segmentation algorithm. The ConvNet was trained and evaluated using a subset of inspiratory thoracic CT scans taken from the COPDGene study. Our method was validated on a separate independent set of the EXACT'09 challenge. We show that our method significantly improves the quality of a given leaky airway segmentation, achieving a higher sensitivity at a low false-positive rate compared to all the state-of-the-art methods that entered in EXACT09, and approaching the performance of the combination of all of them.

Contribution of CT quantified emphysema, air trapping and airway wall thickness on pulmonary function in male smokers with and without COPD.

Emphysema, airway wall thickening and air trapping are associated with chronic obstructive pulmonary disease (COPD). All three can be quantified by computed tomography (CT) of the chest. The goal of the current study is to determine the relative contribution of CT derived parameters on spirometry, lung volume and lung diffusion testing. Emphysema, airway wall thickening and air trapping were quantified automatically on CT in 1,138 male smokers with and without COPD. Emphysema was quantified by the percentage of voxels below -950 Hounsfield Units (HU), airway wall thickness by the square root of wall area for a theoretical airway with 10 mm lumen perimeter (Pi10) and air trapping by the ratio of mean lung density at expiration and inspiration (E/I-ratio). Spirometry, residual volume to total lung capacity (RV/TLC) and diffusion capacity (Kco) were obtained. Standardized regression coefficients (ß) were used to analyze the relative contribution of CT changes to pulmonary function measures. The independent contribution of the three CT measures differed per lung function parameter. For the FEV1 airway wall thickness was the most contributing structural lung change (ß = -0.46), while for the FEV1/FVC this was emphysema (ß = -0.55). For the residual volume (RV) air trapping was most contributing (ß = -0.35). Lung diffusion capacity was most influenced by emphysema (ß = -0.42). In a cohort of smokers with and without COPD the effect of different CT changes varies per lung function measure and therefore emphysema, airway wall thickness and air trapping need to be taken in account.