A Deep Learning Approach for Segmentation, Classification, and Visualization of 3-D High-Frequency Ultrasound Images of Mouse Embryos

Dental bur detection system based on asymmetric double convolution and adaptive feature fusion

This study aims to improve the detection of dental burs, which are often undetected due to their minuscule size, slender profile, and substantial manufacturing output. The present study introduces You Only Look Once-Dental bur (YOLO-DB), an innovative deep learning-driven methodology for the accurate detection and counting of dental burs. A Lightweight Asymmetric Dual Convolution module (LADC) was devised to diminish the detrimental effects of extraneous features on the model's precision, thereby enhancing the feature extraction network. Moreover, to augment the efficiency of feature integration and diminish computational demands, a novel fusion network combining SlimNeck with BiFPN-Concat was introduced, effectively merging superficial spatial details with profound semantic features. A specialized platform was developed for the detection and counting of dental burs, and rigorous experimental assessments were performed. Promising results were achieved. YOLO-DB yielded a Mean Average Precision (mAP@0.5) of 99.3% on the dental bur dataset, with a notable 3.2% increase in mAP@0.5:0.95 and a sustained detection pace of 128 frames per second. The model also achieved a 14.4% reduction in parameter volume and a 17.9% decrease in computational expenditure, while achieving a flawless counting accuracy of 100%. Our approach outperforms current detection algorithms in terms of detection capability and efficiency, presenting a new method for the precise detection and counting of elongated objects such as dental burs.

From beasts to bytes: Revolutionizing zoological research with artificial intelligence

Since the late 2010s, Artificial Intelligence (AI) including machine learning, boosted through deep learning, has boomed as a vital tool to leverage computer vision, natural language processing and speech recognition in revolutionizing zoological research. This review provides an overview of the primary tasks, core models, datasets, and applications of AI in zoological research, including animal classification, resource conservation, behavior, development, genetics and evolution, breeding and health, disease models, and paleontology. Additionally, we explore the challenges and future directions of integrating AI into this field. Based on numerous case studies, this review outlines various avenues for incorporating AI into zoological research and underscores its potential to enhance our understanding of the intricate relationships that exist within the animal kingdom. As we build a bridge between beast and byte realms, this review serves as a resource for envisioning novel AI applications in zoological research that have not yet been explored.

Segmentation of Arm Ultrasound Images in Breast Cancer-Related Lymphedema: A Database and Deep Learning Algorithm

OBJECTIVE

Breast cancer treatment often causes the removal of or damage to lymph nodes of the patient's lymphatic drainage system. This side effect is the origin of Breast Cancer-Related Lymphedema (BCRL), referring to a noticeable increase in excess arm volume. Ultrasound imaging is a preferred modality for the diagnosis and progression monitoring of BCRL because of its low cost, safety, and portability. As the affected and unaffected arms look similar in B-mode ultrasound images, the thickness of the skin, subcutaneous fat, and muscle have been shown to be important biomarkers for this task. The segmentation masks are also helpful in monitoring the longitudinal changes in morphology and mechanical properties of tissue layers.

METHODS

For the first time, a publicly available ultrasound dataset containing the Radio-Frequency (RF) data of 39 subjects and manual segmentation masks by two experts, are provided. Inter- and intra-observer reproducibility studies performed on the segmentation maps show a high Dice Score Coefficient (DSC) of 0.94±0.08 and 0.92±0.06, respectively. Gated Shape Convolutional Neural Network (GSCNN) is modified for precise automatic segmentation of tissue layers, and its generalization performance is improved by the CutMix augmentation strategy.

RESULTS

We got an average DSC of 0.87±0.11 on the test set, which confirms the high performance of the method.

CONCLUSION

Automatic segmentation can pave the way for convenient and accessible staging of BCRL, and our dataset can facilitate development and validation of those methods.

SIGNIFICANCE

Timely diagnosis and treatment of BCRL have crucial importance in preventing irreversible damage.

Applications of Deep Learning Algorithms to Ultrasound Imaging Analysis in Preclinical Studies on In Vivo Animals

BACKGROUND AND AIM

Ultrasound (US) imaging is increasingly preferred over other more invasive modalities in preclinical studies using animal models. However, this technique has some limitations, mainly related to operator dependence. To overcome some of the current drawbacks, sophisticated data processing models are proposed, in particular artificial intelligence models based on deep learning (DL) networks. This systematic review aims to overview the application of DL algorithms in assisting US analysis of images acquired in in vivo preclinical studies on animal models.

METHODS

A literature search was conducted using the Scopus and PubMed databases. Studies published from January 2012 to November 2022 that developed DL models on US images acquired in preclinical/animal experimental scenarios were eligible for inclusion. This review was conducted according to PRISMA guidelines.

RESULTS

Fifty-six studies were enrolled and classified into five groups based on the anatomical district in which the DL models were used. Sixteen studies focused on the cardiovascular system and fourteen on the abdominal organs. Five studies applied DL networks to images of the musculoskeletal system and eight investigations involved the brain. Thirteen papers, grouped under a miscellaneous category, proposed heterogeneous applications adopting DL systems. Our analysis also highlighted that murine models were the most common animals used in in vivo studies applying DL to US imaging.

CONCLUSION

DL techniques show great potential in terms of US images acquired in preclinical studies using animal models. However, in this scenario, these techniques are still in their early stages, and there is room for improvement, such as sample sizes, data preprocessing, and model interpretability.

Longitudinal in Utero Analysis of Engrailed-1 Knockout Mouse Embryonic Phenotypes Using High-Frequency Ultrasound

Large-scale international efforts to generate and analyze loss-of-function mutations in each of the approximately 20,000 protein-encoding gene mutations are ongoing using the "knockout" mouse as a model organism. Because one-third of gene knockouts are expected to result in embryonic lethality, it is important to develop non-invasive in utero imaging methods to detect and monitor mutant phenotypes in mouse embryos. We describe the utility of 3-D high-frequency (40-MHz) ultrasound (HFU) for longitudinal in utero imaging of mouse embryos between embryonic days (E) 11.5 and E14.5, which represent critical stages of brain and organ development. Engrailed-1 knockout (En1-ko) mouse embryos and their normal control littermates were imaged with HFU in 3-D, enabling visualization of morphological phenotypes in the developing brains, limbs and heads of the En1-ko embryos. Recently developed deep learning approaches were used to automatically segment the embryonic brain ventricles and bodies from the 3-D HFU images, allowing quantitative volumetric analyses of the En1-ko brain phenotypes. Taken together, these results show great promise for the application of longitudinal 3-D HFU to analyze knockout mouse embryos in utero.

Deep learning multi-organ segmentation for whole mouse cryo-images including a comparison of 2D and 3D deep networks

Cryo-imaging provided 3D whole-mouse microscopic color anatomy and fluorescence images that enables biotechnology applications (e.g., stem cells and metastatic cancer). In this report, we compared three methods of organ segmentation: 2D U-Net with 2D-slices and 3D U-Net with either 3D-whole-mouse or 3D-patches. We evaluated the brain, thymus, lung, heart, liver, stomach, spleen, left and right kidney, and bladder. Training with 63 mice, 2D-slices had the best performance, with median Dice scores of > 0.9 and median Hausdorff distances of < 1.2 mm in eightfold cross validation for all organs, except bladder, which is a problem organ due to variable filling and poor contrast. Results were comparable to those for a second analyst on the same data. Regression analyses were performed to fit learning curves, which showed that 2D-slices can succeed with fewer samples. Review and editing of 2D-slices segmentation results reduced human operator time from ~ 2-h to ~ 25-min, with reduced inter-observer variability. As demonstrations, we used organ segmentation to evaluate size changes in liver disease and to quantify the distribution of therapeutic mesenchymal stem cells in organs. With a 48-GB GPU, we determined that extra GPU RAM improved the performance of 3D deep learning because we could train at a higher resolution.

Automated segmentation of epidermis in high-frequency ultrasound of pathological skin using a cascade of DeepLab v3+ networks and fuzzy connectedness

This study proposes a novel, fully automated framework for epidermal layer segmentation in different skin diseases based on 75 MHz high-frequency ultrasound (HFUS) image data. A robust epidermis segmentation is a vital first step to detect changes in thickness, shape, and intensity and therefore support diagnosis and treatment monitoring in inflammatory and neoplastic skin lesions. Our framework links deep learning and fuzzy connectedness for image analysis. It consists of a cascade of two DeepLab v3+ models with a ResNet-50 backbone and a fuzzy connectedness analysis module for fine segmentation. Both deep models are pre-trained on the ImageNet dataset and subjected to transfer learning using our HFUS database of 580 images with atopic dermatitis, psoriasis and non-melanocytic skin tumors. The first deep model is used to detect the appropriate region of interest, while the second stands for the main segmentation procedure. We use the softmax layer of the latter twofold to prepare the input data for fuzzy connectedness analysis: as a reservoir of seed points and a direct contribution to the input image. In the experiments, we analyze different configurations of the framework, including region of interest detection, deep model backbones and training loss functions, or fuzzy connectedness analysis with parameter settings. We also use the Dice index and epidermis thickness to compare our results to state-of-the-art approaches. The Dice index of 0.919 yielded by our model over the entire dataset (and exceeding 0.93 in inflammatory diseases) proves its superiority over the other methods.