Search

The recent advances in deep learning have revolutionised computer-aided diagnosis in medical imaging. However, deep learning approaches to unveil their full potential require significant amounts of data, which can be a challenging task in some scientific fields, such as musculoskeletal ultrasound imaging, in which data privacy and security reasons can lead to important limitations in the acquisition and the distribution process of patients’ data. For this reason, different generative methods have been introduced to significantly reduce the required amount of real data by generating synthetic images, almost indistinguishable from the real ones. In this study, the power of the diffusion models is incorporated for the generation of realistic data from a small set of musculoskeletal ultrasound images in four different muscles. Afterwards, the similarity of the generated and real images is assessed with different types of qualitative and quantitative metrics that correspond well with human judgement. In particular, the histograms of pixel intensities of the two sets of images have demonstrated that the two distributions are statistically similar. Additionally, the well-established LPIPS, SSIM, FID, and PSNR metrics have been used to quantify the similarity of these sets of images. The two sets of images have achieved extremely high similarity scores in all these metrics. Subsequently, high-level features are extracted from the two types of images and visualized in a two-dimensional space for inspection of their structure and to identify patterns. From this representation, the two sets of images are hard to distinguish. Finally, we perform a series of experiments to assess the impact of the generated data for training a highly efficient Attention-UNet for the important clinical application of muscle thickness measurement. Our results depict that the synthetic data play a significant role in the model’s final performance and can lead to the improvement of the deep learning systems in musculoskeletal ultrasound.

Summary:. Tranexamic acid (TXA) is a drug commonly used to decrease intraoperative bleeding. Its use in various types of surgery is well defined; however, its use in plastic surgery and even more so, hand surgery is not as well researched. Concurrently, the use of wide-awake local anesthesia no tourniquet (WALANT) procedures has increased in popularity. As the use of WALANT increases, it comes to question whether TXA could be used as an alternative to epinephrine during WALANT procedures. This case involves a 33-year-old woman status-post motor vehicle collision. On examination, the patient sustained transection of the flexor digitorum profundus on both her left middle and ring fingers along with absent sensation to her left middle finger. Due to the surgical preference for vascular monitoring, TXA was used alternatively to epinephrine to control intraoperative bleeding during the tendon repairs. It was observed that TXA allowed for controlled bleeding, preservation of the surgical visual field, successful tendon repair, and postoperative success. The successful use of TXA in this case prompts further investigation as to whether TXA could be used as an alternative to epinephrine in WALANT procedures.

Automatically measuring a muscle’s cross-sectional area is an important application in clinical practice that has been studied extensively in recent years for its ability to assess muscle architecture. Additionally, an adequately segmented cross-sectional area can be used to estimate the echogenicity of the muscle, another valuable parameter correlated with muscle quality. This study assesses state-of-the-art convolutional neural networks and vision transformers for automating this task in a new, large, and diverse database. This database consists of 2005 transverse ultrasound images from four informative muscles for neuromuscular disorders, recorded from 210 subjects of different ages, pathological conditions, and sexes. Regarding the reported results, all of the evaluated deep learning models have achieved near-to-human-level performance. In particular, the manual vs. the automatic measurements of the cross-sectional area exhibit an average discrepancy of less than 38.15 mm2, a significant result demonstrating the feasibility of automating this task. Moreover, the difference in muscle echogenicity estimated from these two readings is only 0.88, another indicator of the proposed method’s success. Furthermore, Bland–Altman analysis of the measurements exhibits no systematic errors since most differences fall between the 95% limits of agreements and the two readings have a 0.97 Pearson’s correlation coefficient (p < 0.001, validation set) with ICC (2, 1) surpassing 0.97, showing the reliability of this approach. Finally, as a supplementary analysis, the texture of the muscle’s visible cross-sectional area was examined using deep learning to investigate whether a classification between healthy subjects and patients with pathological conditions solely from the muscle texture is possible. Our preliminary results indicate that such a task is feasible, but further and more extensive studies are required for more conclusive results.

Evaluation of food microbiome is of major importance since it accounts for the product's organoleptic characteristics and their nutritional value. In this dataset, microbes present in olive samples (‘Kalamata’ variety) stored under modified atmosphere and throughout different time-points of the shelf life of the product are presented, originated after 16S and 18S rRNA sequencing. The different time-points analyzed were: T0 (immediately after packaging), T6 (six months of storage), T12 (12 months of storage) and T18 (six months after the end of shelf life). Sequencing was performed on a MiSeq platform with the MiSeq Reagent Kit v3 (600 cycles). The raw sequence data used for analysis are available in NCBI under the Sequence Read Archive (SRA), with BioProject ID PRJNA688686. Raw reads were analyzed using the QIIME2 pipeline, clustered into Operational Taxonomic Units (OTU) and aligned against SILVA 132 reference database. OTUs are presented in different taxonomic levels for each time-point. These data present valuable information on the microbial communities of table olives, a dynamic niche that affect the final product quality. The data presented are related to the research article “Insights into the evolution of Greek style table olives microbiome stored under modified atmosphere: biochemical implications on the product quality” [1].

The evaluation of the fire impact on building materials is of great scientific and socio-economic importance since fire can result in materials’ chemical and mechanical alterations, which leads to structural stability problems of historical and/or modern construction. This highly increases the cost of rehabilitation interventions. The case study of the Byzantine Monastery of Panagia (Virgin Mary) Varnakova is an example of the fire effect on both historical and newer stone masonries. The Varnakova Monastery is a typical 19th century monastic complex and, during its long history, it has undergone multiple reconstructions after major catastrophic events that have taken place due to its strategic geographical position and its financial and spiritual significance for the region. The last big-scale renovation of the Monastery was conducted between the years 1992 to 2014. However, in January 2017, a devastating fire destroyed the largest part of the monastic cells’ quarter. In this work, a diagnostic study of the different construction phases’ materials comprising the masonries of the monastery cells in their present state is presented. The examination of a series of samples through analytical techniques, such as optical microscopy, X-ray diffraction, thermal analysis, and total immersion tests, along with the use of non-destructive techniques in situ, such as Infra Red Thermography, Digital Microscopy, and Schmidt Hammer Rebound tests, shed light on the preservation state and on the decay of the diverse building materials. In addition, the impact of the fire on their properties was investigated. The results reveal the diversity of the materials used in the historical masonries throughout the centuries, while the combination of analytical and non-destructive techniques demonstrates the damages induced by the fire.

Despite the success of hand-crafted features in computer visioning for many years, nowadays, this has been replaced by end-to-end learnable features that are extracted from deep convolutional neural networks (CNNs). Whilst CNNs can learn robust features directly from image pixels, they require large amounts of samples and extreme augmentations. On the contrary, hand-crafted features, like SIFT, exhibit several interesting properties as they can provide local rotation invariance. In this work, a novel scheme combining the strengths of SIFT descriptors with CNNs, namely SIFT-CNN, is presented. Given a single-channel image, one SIFT descriptor is computed for every pixel, and thus, every pixel is represented as an M-dimensional histogram, which ultimately results in an M-channel image. Thus, the SIFT image is generated from the SIFT descriptors for all the pixels in a single-channel image, while at the same time, the original spatial size is preserved. Next, a CNN is trained to utilize these M-channel images as inputs by operating directly on the multiscale SIFT images with the regular convolution processes. Since these images incorporate spatial relations between the histograms of the SIFT descriptors, the CNN is guided to learn features from local gradient information of images that otherwise can be neglected. In this manner, the SIFT-CNN implicitly acquires a local rotation invariance property, which is desired for problems where local areas within the image can be rotated without affecting the overall classification result of the respective image. Some of these problems refer to indirect immunofluorescence (IIF) cell image classification, ground-based all-sky image-cloud classification and human lip-reading classification. The results for the popular datasets related to the three different aforementioned problems indicate that the proposed SIFT-CNN can improve the performance and surpasses the corresponding CNNs trained directly on pixel values in various challenging tasks due to its robustness in local rotations. Our findings highlight the importance of the input image representation in the overall efficiency of a data-driven system.

Automatically delineating the deep and superficial aponeurosis of the skeletal muscles from ultrasound images is important in many aspects of the clinical routine. In particular, finding muscle parameters, such as thickness, fascicle length or pennation angle, is a time-consuming clinical task requiring both human labour and specialised knowledge. In this study, a multi-step solution for automating these tasks is presented. A process to effortlessly extract the aponeurosis for automatically measuring the muscle thickness has been introduced as a first step. This process consists mainly of three parts. In the first part, the Attention UNet has been incorporated to automatically delineate the boundaries of the studied muscles. Afterwards, a specialised post-processing algorithm was utilised to improve (and correct) the segmentation results. Lastly, the calculation of the muscle thickness was performed. The proposed method has achieved similar to a human-level performance. In particular, the overall discrepancy between the automatic and the manual muscle thickness measurements was equal to 0.4 mm, a significant result that demonstrates the feasibility of automating this task. In the second step of the proposed methodology, the fascicle’s length and pennation angle are extracted through an unsupervised pipeline. Initially, filtering is applied to the ultrasound images to further distinguish the tissues from the other muscle structures. Later, the well-known K-Means algorithm is used to isolate them successfully. As the last step, the dominant angle of the segmented muscle tissues is reported and compared with manual measurements. The proposed pipeline is showing very promising results in the evaluated dataset. Specifically, in the calculation of the pennation angle, the overall discrepancy between the automatic and the manual measurements was less than 2.22° (degrees), once more comparable with the human-level performance. Finally, regarding the fascicle length measurements, the results were divided based on the muscle properties. In the muscles where a large portion (or all) of the fascicles are located between the upper and lower aponeuroses, the proposed pipeline exhibits superb performance; otherwise, overall accuracy deteriorates due to errors caused by the trigonometric approximations needed for the length calculation.

In this work, we propose a mechanism for knowledge transfer between Convolutional Neural Networks via the geometric regularization of local features produced by the activations of convolutional layers. We formulate appropriate loss functions, driving a “student” model to adapt such that its local features exhibit similar geometrical characteristics to those of an “instructor” model, at corresponding layers. The investigated functions, inspired by manifold-to-manifold distance measures, are designed to compare the neighboring information inside the feature space of the involved activations without any restrictions in the features’ dimensionality, thus enabling knowledge transfer between different architectures. Experimental evidence demonstrates that the proposed technique is effective in different settings, including knowledge-transfer to smaller models, transfer between different deep architectures and harnessing knowledge from external data, producing models with increased accuracy compared to a typical training. Furthermore, results indicate that the presented method can work synergistically with methods such as knowledge distillation, further increasing the accuracy of the trained models. Finally, experiments on training with limited data show that a combined regularization scheme can achieve the same generalization as a non-regularized training with 50% of the data in the CIFAR-10 classification task.

Background. P63 immunostaining has been considered as potential prognostic factor in laryngeal cancer. Considering that P63 is mainly nuclear stain, a possible correlation between the texture of P63-stained nuclei and the tumor’s grade could be of value to diagnosis, since this may be related to biologic information imprinted as texture on P63 expressed nuclei. Objective. To investigate the association between P63 stained nuclei and histologic grade in laryngeal tumor lesions. Methods. Biopsy specimens from laryngeal tumour lesions of 55 patients diagnosed with laryngeal squamous cell carcinomas were immunohistochemically (IHC) stained for P63 expression. Four images were digitized from each patient’s IHC specimens. P63 positively expressed nuclei were identified, the percentage of P63 expressed nuclei was computed, and 118 textural, morphological, shape, and architectural features were calculated from each one of the 55 laryngeal lesions. Data were split into the low grade (21 grade I lesions) and high grade (34 grade II and grade III lesions) classes for statistical analysis. Results. With advancing grade, P63 expression decreased, P63 stained nuclei appeared of lower image intensity, more inhomogeneous, of higher local contrast, contained smaller randomly distributed dissimilar structures and had irregular shape. Conclusion. P63 expressed nuclei contain important information related to histologic grade.

A human-centered approach to image database organization is presented in this study. The management of a generic image database is pursued using a standard psychophysical experimental procedure followed by a well-suited data analysis methodology that is based on simple geometrical concepts. The end result is a cognitive discriminative biplot, which is a visualization of the intrinsic organization of the image database best reflecting the user's perception. The discriminating power of the introduced cognitive biplot constitutes an appealing tool for image retrieval and a flexible interface for visual data mining tasks. These ideas were evaluated in two ways. First, the separability of semantically distinct image classes was measured according to their reduced representations on the biplot. Then, a nearest-neighbor retrieval scheme was run on the emerged low-dimensional terrain to measure the suitability of the biplot for performing content-based image retrieval (CBIR). The achieved organization performance when compared with the performance of a contemporary system was found superior. This promoted the further discussion of packing these ideas into a realizable algorithmic procedure for an efficient and effective personalized CBIR system.