Learnable DoG convolutional filters for microcalcification detection.

Difference of Gaussians (DoG) convolutional filters are one of the earliest image processing methods employed for detecting microcalcifications on mammogram images before machine and deep learning methods became widespread. DoG is a blob enhancement filter that consists in subtracting one Gaussian-smoothed version of an image from another less Gaussian-smoothed version of the same image. Smoothing with a Gaussian kernel suppresses high-frequency spatial information, thus DoG can be regarded as a band-pass filter. However, due to their small size and overimposed breast tissue, microcalcifications vary greatly in contrast-to-noise ratio and sharpness. This makes it difficult to find a single DoG configuration that enhances all microcalcifications. In this work, we propose a convolutional network, named DoG-MCNet, where the first layer automatically learns a bank of DoG filters parameterized by their associated standard deviations. We experimentally show that when employed for microcalcification detection, our DoG layer acts as a learnable bank of band-pass preprocessing filters and improves detection performance by 4.86% AUFROC over baseline MCNet and 1.53% AUFROC over state-of-the-art multicontext ensemble of CNNs.

Convolutional Networks and Transformers for Mammography Classification: An Experimental Study.

Convolutional Neural Networks (CNN) have received a large share of research in mammography image analysis due to their capability of extracting hierarchical features directly from raw data. Recently, Vision Transformers are emerging as viable alternative to CNNs in medical imaging, in some cases performing on par or better than their convolutional counterparts. In this work, we conduct an extensive experimental study to compare the most recent CNN and Vision Transformer architectures for whole mammograms classification. We selected, trained and tested 33 different models, 19 convolutional- and 14 transformer-based, on the largest publicly available mammography image database OMI-DB. We also performed an analysis of the performance at eight different image resolutions and considering all the individual lesion categories in isolation (masses, calcifications, focal asymmetries, architectural distortions). Our findings confirm the potential of visual transformers, which performed on par with traditional CNNs like ResNet, but at the same time show a superiority of modern convolutional networks like EfficientNet.

CT- and MRI-Based 3D Reconstruction of Knee Joint to Assess Cartilage and Bone.

For the observation of human joint cartilage, X-ray, computed tomography (CT) or magnetic resonance imaging (MRI) are the main diagnostic tools to evaluate pathologies or traumas. The current work introduces a set of novel measurements and 3D features based on MRI and CT data of the knee joint, used to reconstruct bone and cartilages and to assess cartilage condition from a new perspective. Forty-seven subjects presenting a degenerative disease, a traumatic injury or no symptoms or trauma were recruited in this study and scanned using CT and MRI. Using medical imaging software, the bone and cartilage of the knee joint were segmented and 3D reconstructed. Several features such as cartilage density, volume and surface were extracted. Moreover, an investigation was carried out on the distribution of cartilage thickness and curvature analysis to identify new markers of cartilage condition. All the extracted features were used with advanced statistics tools and machine learning to test the ability of our model to predict cartilage conditions. This work is a first step towards the development of a new gold standard of cartilage assessment based on 3D measurements.

Trends in IoT based solutions for health care: Moving AI to the edge.

In recent times, we assist to an ever growing diffusion of smart medical sensors and Internet of things devices that are heavily changing the way healthcare is approached worldwide. In this context, a combination of Cloud and IoT architectures is often exploited to make smart healthcare systems capable of supporting near realtime applications when processing and performing Artificial Intelligence on the huge amount of data produced by wearable sensor networks. Anyway, the response time and the availability of cloud based systems, together with security and privacy, still represent critical issues that prevents Internet of Medical Things (IoMT) devices and architectures from being a reliable and effective solution to the aim. Lately, there is a growing interest towards architectures and approaches that exploit Edge and Fog computing as an answer to compensate the weaknesses of the cloud. In this paper, we propose a short review about the general use of IoT solutions in health care, starting from early health monitoring solutions from wearable sensors up to a discussion about the latest trends in fog/edge computing for smart health.

Addressing class imbalance in deep learning for small lesion detection on medical images.

Deep learning methods utilizing Convolutional Neural Networks (CNNs) have led to dramatic advances in automated understanding of medical images. However, in many medical image classification tasks, lesions occupy only a few pixels of the image. This results in a significant class imbalance between lesion and background. From recent literature, it is known that class imbalance may negatively affect the performance of CNN classification. However, very few research exists in the context of lesion detection. In this work, we propose a two-stage deep learning framework able to deal with the high class imbalance encountered during training of small lesion detectors. First, we train a deep cascade (DC) of long sequences of decision trees with an algorithm designed to handle unbalanced data that also drastically reduces the number of background samples reaching the final stage. The remaining samples are fed to a CNN, whose training benefits from both rebalance and hard mining done by the DC. We evaluated DC-CNN on two severely unbalanced classification problems: microcalcification detection and microaneurysm detection. In both cases, DC-CNN outperformed the CNNs trained with commonly used methods for addressing class imbalance such as oversampling, undersampling, hard mining, cost sensitive learning, and one-class classification. The DC-CNN was also ∼10x faster than CNN at test time.

Improving the Automated Detection of Calcifications Using Adaptive Variance Stabilization.

In this paper, we analyze how stabilizing the variance of intensity-dependent quantum noise in digital mammograms can significantly improve the computerized detection of microcalcifications (MCs). These lesions appear on mammograms as tiny deposits of calcium smaller than 20 pixels in diameter. At this scale, high frequency image noise is dominated by quantum noise, which in raw mammograms can be described with a square-root noise model. Under this assumption, we derive an adaptive variance stabilizing transform (VST) that stabilizes the noise to unitary standard deviation in all the images. This is achieved by estimating the noise characteristics from the image at hand. We tested the adaptive VST as a preprocessing stage for four existing computerized MC detection methods on three data sets acquired with mammographic units from different manufacturers. In all the test cases considered, MC detection performance on transformed mammograms was statistically significantly higher than on unprocessed mammograms. Results were also superior in comparison with a "fixed" (nonparametric) VST previously proposed for digital mammograms.

Improving computer-aided detection assistance in breast cancer screening by removal of obviously false-positive findings.

PURPOSE: Computer-aided detection (CADe) systems for mammography screening still mark many false positives. This can cause radiologists to lose confidence in CADe, especially when many false positives are obviously not suspicious to them. In this study, we focus on obvious false positives generated by microcalcification detection algorithms. METHODS: We aim at reducing the number of obvious false-positive findings by adding an additional step in the detection method. In this step, a multiclass machine learning method is implemented in which dedicated classifiers learn to recognize the patterns of obvious false-positive subtypes that occur most frequently. The method is compared to a conventional two-class approach, where all false-positive subtypes are grouped together in one class, and to the baseline CADe system without the new false-positive removal step. The methods are evaluated on an independent dataset containing 1,542 screening examinations of which 80 examinations contain malignant microcalcifications. RESULTS: Analysis showed that the multiclass approach yielded a significantly higher sensitivity compared to the other two methods (P < 0.0002). At one obvious false positive per 100 images, the baseline CADe system detected 61% of the malignant examinations, while the systems with the two-class and multiclass false-positive reduction step detected 73% and 83%, respectively. CONCLUSIONS: Our study showed that by adding the proposed method to a CADe system, the number of obvious false positives can decrease significantly (P < 0.0002).

On linear combinations of dichotomizers for maximizing the area under the ROC curve.

In this paper, we propose a method for the linear combination of several dichotomizers aimed at maximizing the area under the receiver operating characteristic (ROC) curve of the resulting classification system. This is particularly suited for real applications where it is difficult to exactly determine the key parameters such as costs and priors. In such cases, the accuracy is not adequate in measuring the quality of a classification system, while the ROC analysis provides the right tools for an appropriate assessment of the classification performance. The proposed approach revealed to be particularly effective with respect to other widespread combination rules both on artificial and real applications.

A computer-aided detection system for clustered microcalcifications.

OBJECTIVE: The aim of this paper is to describe a novel system for computer-aided detection of clusters of microcalcifications on digital mammograms. METHODS AND MATERIAL: Mammograms are first segmented by means of a tree-structured Markov random field algorithm that extracts the elementary homogeneous regions of interest. An analysis of such regions is then performed by means of a two-stage, coarse-to-fine classification based on both heuristic rules and classifier combination. In this phase, we avoid taking a decision on the single microcalcifications and forward it to the successive phase of clustering realized through a sequential approach. RESULTS: The system has been tested on a publicly available database of mammograms and compared with previous approaches. The obtained results show that the system is very effective, especially in terms of sensitivity. CONCLUSIONS: The proposed approach exhibits some remarkable advantages both in segmentation and classification phases. The segmentation phase employs an image model that reduces the computational burden, preserving the small details in the image through an adaptive local estimation of all model parameters. The classification stage combines the results of the classifiers focused on the single microcalcification and the cluster as a whole. Such an approach makes a detection system particularly effective and robust with respect to the large variations exhibited by the clusters of microcalcifications.