Line-field confocal optical coherence tomography for in vivo visualization of morphological characteristics of squamous cell carcinoma in a murine model.

OBJECTIVES: Non-invasive imaging with line-field confocal optical coherence tomography (LC-OCT) can support the diagnosis of squamous cell carcinoma (SCC) through visualization of morphological characteristics specific to skin cancer. We aimed to visualize prominent morphological characteristics of SCC using LC-OCT in a well-established murine SCC model. MATERIALS AND METHODS: Nine hairless mice were exposed to ultraviolet radiation three times weekly for 9 months to induce SCC development. Visible SCC tumors (n = 9) were imaged with LC-OCT and the presence of 10 well-described morphological characteristics of SCC were evaluated in the scans by two physicians with adjudication by a third. RESULTS: Overall, murine morphological characteristics resembled corresponding features previously reported in human SCCs. Interrupted dermal-epidermal junction occurred in 100% of tumors. In epidermis, the most frequently observed characteristics were severe epidermal dysplasia (100%) and tumor budding (89%). Common dermal characteristics included broad strands (100%) and collagen alterations (78%). CONCLUSION: LC-OCT imaging can be used to non-invasively visualize morphological characteristics specific to SCC in an in vivo preclinical model.

Preprocessing of Iris Images for BSIF-Based Biometric Systems: Binary Detected Edges and Iris Unwrapping.

This work presents a novel approach to enhancing iris recognition systems through a two-module approach focusing on low-level image preprocessing techniques and advanced feature extraction. The primary contributions of this paper include: (i) the development of a robust preprocessing module utilizing the Canny algorithm for edge detection and the circle-based Hough transform for precise iris extraction, and (ii) the implementation of Binary Statistical Image Features (BSIF) with domain-specific filters trained on iris-specific data for improved biometric identification. By combining these advanced image preprocessing techniques, the proposed method addresses key challenges in iris recognition, such as occlusions, varying pigmentation, and textural diversity. Experimental results on the Human-inspired Domain-specific Binarized Image Features (HDBIF) Dataset, consisting of 1892 iris images, confirm the significant enhancements achieved. Moreover, this paper offers a comprehensive and reproducible research framework by providing source codes and access to the testing database through the Notre Dame University dataset website, thereby facilitating further application and study. Future research will focus on exploring adaptive algorithms and integrating machine learning techniques to improve performance across diverse and unpredictable real-world scenarios.

MRI image features with an evident relation to low back pain: a narrative review.

PURPOSE: Low back pain (LBP) is one of the most prevalent health condition worldwide and responsible for the most years lived with disability, yet the etiology is often unknown. Magnetic resonance imaging (MRI) is frequently used for treatment decision even though it is often inconclusive. There are many different image features that could relate to low back pain. Conversely, multiple etiologies do relate to spinal degeneration but do not actually cause the perceived pain. This narrative review provides an overview of all possible relevant features visible on MRI images and determines their relation to LBP. METHODS: We conducted a separate literature search per image feature. All included studies were scored using the GRADE guidelines. Based on the reported results per feature an evidence agreement (EA) score was provided, enabling us to compare the collected evidence of separate image features. The various relations between MRI features and their associated pain mechanisms were evaluated to provide a list of features that are related to LBP. RESULTS: All searches combined generated a total of 4472 hits of which 31 articles were included. Features were divided into five different categories:'discogenic', 'neuropathic','osseous', 'facetogenic', and'paraspinal', and discussed separately. CONCLUSION: Our research suggests that type I Modic changes, disc degeneration, endplate defects, disc herniation, spinal canal stenosis, nerve compression, and muscle fat infiltration have the highest probability to be related to LBP. These can be used to improve clinical decision-making for patients with LBP based on MRI.

A Comprehensive Landscape of Imaging Feature-Associated RNA Expression Profiles in Human Breast Tissue.

The expression abundance of transcripts in nondiseased breast tissue varies among individuals. The association study of genotypes and imaging phenotypes may help us to understand this individual variation. Since existing reports mainly focus on tumors or lesion areas, the heterogeneity of pathological image features and their correlations with RNA expression profiles for nondiseased tissue are not clear. The aim of this study is to discover the association between the nucleus features and the transcriptome-wide RNAs. We analyzed both microscopic histology images and RNA-sequencing data of 456 breast tissues from the Genotype-Tissue Expression (GTEx) project and constructed an automatic computational framework. We classified all samples into four clusters based on their nucleus morphological features and discovered feature-specific gene sets. The biological pathway analysis was performed on each gene set. The proposed framework evaluates the morphological characteristics of the cell nucleus quantitatively and identifies the associated genes. We found image features that capture population variation in breast tissue associated with RNA expressions, suggesting that the variation in expression pattern affects population variation in the morphological traits of breast tissue. This study provides a comprehensive transcriptome-wide view of imaging-feature-specific RNA expression for healthy breast tissue. Such a framework could also be used for understanding the connection between RNA expression and morphology in other tissues and organs. Pathway analysis indicated that the gene sets we identified were involved in specific biological processes, such as immune processes.

Fusion extracted features from deep learning for identification of multiple positioning errors in dental panoramic imaging.

BACKGROUND: Dental panoramic imaging plays a pivotal role in dentistry for diagnosis and treatment planning. However, correctly positioning patients can be challenging for technicians due to the complexity of the imaging equipment and variations in patient anatomy, leading to positioning errors. These errors can compromise image quality and potentially result in misdiagnoses. OBJECTIVE: This research aims to develop and validate a deep learning model capable of accurately and efficiently identifying multiple positioning errors in dental panoramic imaging. METHODS AND MATERIALS: This retrospective study used 552 panoramic images selected from a hospital Picture Archiving and Communication System (PACS). We defined six types of errors (E1-E6) namely, (1) slumped position, (2) chin tipped low, (3) open lip, (4) head turned to one side, (5) head tilted to one side, and (6) tongue against the palate. First, six Convolutional Neural Network (CNN) models were employed to extract image features, which were then fused using transfer learning. Next, a Support Vector Machine (SVM) was applied to create a classifier for multiple positioning errors, using the fused image features. Finally, the classifier performance was evaluated using 3 indices of precision, recall rate, and accuracy. RESULTS: Experimental results show that the fusion of image features with six binary SVM classifiers yielded high accuracy, recall rates, and precision. Specifically, the classifier achieved an accuracy of 0.832 for identifying multiple positioning errors. CONCLUSIONS: This study demonstrates that six SVM classifiers effectively identify multiple positioning errors in dental panoramic imaging. The fusion of extracted image features and the employment of SVM classifiers improve diagnostic precision, suggesting potential enhancements in dental imaging efficiency and diagnostic accuracy. Future research should consider larger datasets and explore real-time clinical application.

Identification of Solid and Liquid Materials Using Acoustic Signals and Frequency-Graph Features.

Material identification is playing an increasingly important role in various sectors such as industry, petrochemical, mining, and in our daily lives. In recent years, material identification has been utilized for security checks, waste sorting, etc. However, current methods for identifying materials require direct contact with the target and specialized equipment that can be costly, bulky, and not easily portable. Past proposals for addressing this limitation relied on non-contact material identification methods, such as Wi-Fi-based and radar-based material identification methods, which can identify materials with high accuracy without physical contact; however, they are not easily integrated into portable devices. This paper introduces a novel non-contact material identification based on acoustic signals. Different from previous work, our design leverages the built-in microphone and speaker of smartphones as the transceiver to identify target materials. The fundamental idea of our design is that acoustic signals, when propagated through different materials, reach the receiver via multiple paths, producing distinct multipath profiles. These profiles can serve as fingerprints for material identification. We captured and extracted them using acoustic signals, calculated channel impulse response (CIR) measurements, and then extracted image features from the time-frequency domain feature graphs, including histogram of oriented gradient (HOG) and gray-level co-occurrence matrix (GLCM) image features. Furthermore, we adopted the error-correcting output code (ECOC) learning method combined with the majority voting method to identify target materials. We built a prototype for this paper using three mobile phones based on the Android platform. The results from three different solid and liquid materials in varied multipath environments reveal that our design can achieve average identification accuracies of 90% and 97%.

Contactless Palmprint Recognition Using Binarized Statistical Image Features-Based Multiresolution Analysis.

In recent years, palmprint recognition has gained increased interest and has been a focus of significant research as a trustworthy personal identification method. The performance of any palmprint recognition system mainly depends on the effectiveness of the utilized feature extraction approach. In this paper, we propose a three-step approach to address the challenging problem of contactless palmprint recognition: (1) a pre-processing, based on median filtering and contrast limited adaptive histogram equalization (CLAHE), is used to remove potential noise and equalize the images' lighting; (2) a multiresolution analysis is applied to extract binarized statistical image features (BSIF) at several discrete wavelet transform (DWT) resolutions; (3) a classification stage is performed to categorize the extracted features into the corresponding class using a K-nearest neighbors (K-NN)-based classifier. The feature extraction strategy is the main contribution of this work; we used the multiresolution analysis to extract the pertinent information from several image resolutions as an alternative to the classical method based on multi-patch decomposition. The proposed approach was thoroughly assessed using two contactless palmprint databases: the Indian Institute of Technology-Delhi (IITD) and the Chinese Academy of Sciences Institute of Automatisation (CASIA). The results are impressive compared to the current state-of-the-art methods: the Rank-1 recognition rates are 98.77% and 98.10% for the IITD and CASIA databases, respectively.

Understanding Dilated Mathematical Relationship between Image Features and the Convolutional Neural Network's Learnt Parameters.

Deep learning, in general, was built on input data transformation and presentation, model training with parameter tuning, and recognition of new observations using the trained model. However, this came with a high computation cost due to the extensive input database and the length of time required in training. Despite the model learning its parameters from the transformed input data, no direct research has been conducted to investigate the mathematical relationship between the transformed information (i.e., features, excitation) and the model's learnt parameters (i.e., weights). This research aims to explore a mathematical relationship between the input excitations and the weights of a trained convolutional neural network. The objective is to investigate three aspects of this assumed feature-weight relationship: (1) the mathematical relationship between the training input images' features and the model's learnt parameters, (2) the mathematical relationship between the images' features of a separate test dataset and a trained model's learnt parameters, and (3) the mathematical relationship between the difference of training and testing images' features and the model's learnt parameters with a separate test dataset. The paper empirically demonstrated the existence of this mathematical relationship between the test image features and the model's learnt weights by the ANOVA analysis.

The ultrafast, high-pitch turbo FLASH mode of third-generation dual-source CT: Effect of different pitch and corresponding SFOV on image quality in a phantom study.

PURPOSE: To investigate the effect of different pitches and corresponding scan fields of view (SFOVs) on the image quality in the ultrafast, high-pitch turbo FLASH mode of the third-generation dual-source CT using an anthropomorphic phantom. METHODS: The phantom was scanned using the ultrafast, high-pitch turbo FLASH protocols of the third-generation dual-source CT with the different pitches and corresponding SFOVs (pitches: 1.55 to 3.2 with increments of 0.1, SFOVs: 50 cm to 35.4 cm). The objective parameters such as the CT number, image noises, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and artifacts index (AI), and image features from the head, chest, and abdomen were compared between the CT images with a pitch of 1.55 and SFOV of Ø 50 cm and a pitch of 3.2 and SFOV of Ø 35.4 cm. Then, the 18 series of CT images of the head, chest, and abdomen were evaluated by three radiologists independently. RESULTS: The differences in the CT numbers were not statically significant between the CT images with a pitch of 1.55 and SFOV of Ø 50 cm and a pitch of 3.2 and SFOV of Ø 35.4 cm from most body parts and potential combinations (p > 0.05), Most of the image noises and the AI from the images with the pitch of 1.55 were significantly lower than those with the pitch of 3.2 (p < 0.05), and the SNR and CNR from the images with the pitch of 1.55 were higher than those with the pitch of 3.2. There were significant differences in the first-order features and texture features of the head (59.3%, 28.3%), chest (66%, 35.7%), and abdomen (71.6%, 64.7%) (p < 0.05). The subjective image quality was excellent when the pitch was less than 2.0 and gradually decreased with the increasing pitch. In addition, the image quality decreased significantly when the pitch was higher than 3.0 (all k≥0.69), especially in the head and chest. CONCLUSIONS: In the ultrafast, high-pitch turbo FLASH mode of the third-generation DSCT, increasing the pitch and lowering the corresponding SFOV will change the image features and cause more artifacts degrading the image quality. Specific to the clinical needs, decreasing the pitch not only can expand the SFOV but also can improve the image quality.

Sensing and Rendering Method of 2-Dimensional Haptic Texture.

Nowadays, touchscreens have been used worldwide. However, most of them lack realistic haptic feedback. Several haptic feedback devices employ one-dimensional vibration only. We aim at a novel rendering method for direction-controlled 2-dimensional vibration display to present texture information. This paper proposed a rendering method of texture information that enables lateral-force-based 2-dimensional vibration in the X and Y-axis. Moreover, we proposed combining AKAZE image feature information of the textures to improve the fidelity for larger periodic textures. We held experiments to evaluate the fidelity of the proposed method. The result shows that the proposed method has higher fidelity in presenting randomized textures and large periodic textures than the conventional method.

Multi-Block Color-Binarized Statistical Images for Single-Sample Face Recognition.

Single-Sample Face Recognition (SSFR) is a computer vision challenge. In this scenario, there is only one example from each individual on which to train the system, making it difficult to identify persons in unconstrained environments, mainly when dealing with changes in facial expression, posture, lighting, and occlusion. This paper discusses the relevance of an original method for SSFR, called Multi-Block Color-Binarized Statistical Image Features (MB-C-BSIF), which exploits several kinds of features, namely, local, regional, global, and textured-color characteristics. First, the MB-C-BSIF method decomposes a facial image into three channels (e.g., red, green, and blue), then it divides each channel into equal non-overlapping blocks to select the local facial characteristics that are consequently employed in the classification phase. Finally, the identity is determined by calculating the similarities among the characteristic vectors adopting a distance measurement of the K-nearest neighbors (K-NN) classifier. Extensive experiments on several subsets of the unconstrained Alex and Robert (AR) and Labeled Faces in the Wild (LFW) databases show that the MB-C-BSIF achieves superior and competitive results in unconstrained situations when compared to current state-of-the-art methods, especially when dealing with changes in facial expression, lighting, and occlusion. The average classification accuracies are 96.17% and 99% for the AR database with two specific protocols (i.e., Protocols I and II, respectively), and 38.01% for the challenging LFW database. These performances are clearly superior to those obtained by state-of-the-art methods. Furthermore, the proposed method uses algorithms based only on simple and elementary image processing operations that do not imply higher computational costs as in holistic, sparse or deep learning methods, making it ideal for real-time identification.

[Fusion of spectrum and image features to identify Glycyrrhizae Radix et Rhizoma from different origins based on hyperspectral imaging technology].

To identify Glycyrrhizae Radix et Rhizoma from different geographical origins, spectrum and image features were extracted from visible and near-infrared(VNIR, 435-1 042 nm) and short-wave infrared(SWIR, 898-1 751 nm) ranges based on hyperspectral imaging technology. The spectral features of Glycyrrhizae Radix et Rhizoma samples were extracted from hyperspectral data and denoised by a variety of pre-processing methods. The classification models were established by using Partial Least Squares Discriminate Analysis(PLS-DA), Support Vector Classification(SVC) and Random Forest(RF). Meanwhile, Gray-Level Co-occurrence matrix(GLCM) was employed to extract textural variables. The spectrum and image data were implemented from three dimensions, including VNIR and SWIR fusion, spectrum and image fusion, and comprehensive data fusion. The results indicated that the spectrum in SWIR range performed better classification accuracy than VNIR range. Compared with other four pre-processing methods, the second derivative method based on Savitzky-Golay(SG) smoothing exhibited the best performance, and the classification accuracy of PLS-DA and SVC models were 93.40% and 94.11%, separately. In addition, the PLS-DA model was superior to SVC and RF models in terms of classification accuracy and model generalization capability, which were evaluated by confusion matrix and receiver operating characteristic curve(ROC). Comprehensive data fusion on SPA bands achieved a classification accuracy of 94.82% with only 28 bands. As a result, this approach not only greatly improved the classification efficiency but also maintained its accuracy. The hyperspectral imaging system, a non-invasively, intuitively and quickly identify technology, could effectively distinguish Glycyrrhizae Radix et Rhizoma samples from different origins.

Clinical characteristics of pulmonary alveolar proteinosis. / 肺泡蛋白沉积症的临床特征.

OBJECTIVES: Pulmonary alveolar proteinosis (PAP) is a rare disease with non-specific and various clinical manifestations, often leading to misdiagnosis. This study aims to raise the awareness of this disease via summarizing the clinical characteristics, diagnosis, and therapy of PAP. METHODS: We retrospectively analyzed clinical data of 25 hospitalized cases of PAP during 2008 and 2019 in the Department of Respiratory and Critical Care Medicine of the Second Xiangya Hospital of Central South University. RESULTS: Cough with unkown reason and dyspnea were common clinical manifastations of PAP. Five patients had a history of occupational inhalational exposure. Sixteen patients had typical image features including ground-glass opacification of alveolar spaces and thickening of the interlobular and intralobular septa, in typical shapes called crazy-paving and geographic pattern. Fourteen patients underwent pulmonary function tests, revealing a reduction in the diffusing capacity for carbon monoxide. The positive rate of transbronchial biopsy was 95%. Five patients received the whole lung lavage and the symptoms and imaging fcauters significantly relieved after five-years follow-up. CONCLUSIONS: PAP is characterized by radiographic pattern and pathology. Transbronchial lung biopsy is effective to make diagnosis of PAP. The whole lung lavage remains a efficient therapy.

[Control study of chest CT imaging features of aluminosis and silicosis patients].

Objective: To understand the chest CT features of aluminosis caused by alumina and to improve the understanding of the imaging findings of alumina pneumoconiosis. Methods: The chest CT findings of 17 cases of alumina-induced pneumoconiosis and 30 cases of silicosis (the control group) diagnosed in Zibo Occupational Disease Prevention Hospital from April 2015 to July 2020 were analyzed retrospectively. The characteristics of fibrosis of the two kinds of pneumoconiosis and the incidence of size, density, distribution, tractive bronchiectasis, pleural thickening and interlobular septal thickening of pneumoconiosis nodules were compared. Results: Alumina pneumoconiosis showed nodules with thickened interlobular septal of 66.67% (12/18) , honeycomb lung of 22.22% (4/18) , ground glass shadow of 61.11% (11/18) , simple nodules of 11.11% (2/18) , and no fusion mass. In the control group, the long-line fibrosis of nodules with thickened interlobular septal were 16.67% (5/30) , 6.67% (2/30) with honeycomb lung and ground glass density shadow, 23.33% (7/30) with fusion mass and 53.33% (16/30) with simple nodule. There were significant differences in CT findings of nodules with thickened interlobular septal, ground glass density shadow, fused mass and simple nodules between the two groups (P<0.05) . The interstitial beaded nodules were seen in 18 cases of alumina pneumoconiosis, 50.00% (9/18) of them were beaded nodules, 61.33% (46/75) of low density nodules and 38.89% (7/18) of central lobular nodules were seen in alumina pneumoconiosis. The average width of nodules was (1.29±0.38) mm. Central lobular nodules were seen in all 30 cases of silicosis, 10.00% (3/30) were mainly beaded nodules, low density nodules were 36.29% (90/248) , and the average width diameter of nodules was (1.85±0.58) mm. There were significant differences between the two groups (P<0.05) . Alumina pneumoconiosis was often accompanied by traction bronchiectasis, pleural thickening and interlobular septal thickening (11, 18, 17 cases, 61.11%, 100.00%, 94.44%) , compared with the control group (9, 18, 18 cases, 30.00%, 60.00%, 60.00%) . The differences were statistically significant (P<0.05) . The maximum CT value of noncalcified mediastinal lymphnodes in alumina pneumoconiosis was (103.43±26.33) HU, which was higher than that of the control group[ (75.22±16.70) HU], and the difference was statistically significant (P<0.05) . Conclusion: Alumina pneumoconiosis chest CT shows slightly low-density beaded nodules, thickened interlobular septal, and pulmonary interstitial fibrosis of ground-glass shadows, mostly combines with stretched bronchiectasis, thickened pleura, and mediastinum increased lymph node density.

Automated approach for estimation of grade groups for prostate cancer based on histological image feature analysis.

BACKGROUND: There is a low reproducibility of the Gleason scores that determine the grade group of prostate cancer given the intra- and interobserver variability among pathologists. This study aimed to develop an automated approach for estimating prostate cancer grade groups based on features obtained from histological image analysis. METHODS: Fifty-nine patients who underwent radical prostatectomy were selected under the approval of the institutional review board of our university hospital. For estimation, we followed the grade group criteria provided by the International Society of Urological Pathology in 2014. One hundred eight specimen slides obtained from the patients were digitized to extract 110 regions of interest (ROI) from hematoxylin and eosin-stained histological images using a digital whole slide scanner at ×20 magnification with a pixel size of 0.4 µm. Each color pixel value in the ROI was decomposed into six intensities corresponding to the RGB (red, green, and blue) and HSV (hue, saturation, and value) color models. Image features were extracted by histological image analysis, obtaining 54 features from the ROI based on histogram and texture analyses in the six types of decomposed histological images. Then, 40 representative features were selected from the 324 histological image features based on statistically significant differences (P < .05) between the mean image feature values for high (≥3, Gleason score ≥4 + 3) and low (≤2, Gleason score ≤3 + 4) grade groups. The relationship between grade groups and the most representative image feature (ie, complexity) was approximated using regression to estimate real-number grade groups defined by continuous numerical grading. Finally, the grade groups were expressed as the conventional grade groups (ie, integers from 1 to 5) using a piecewise step function. RESULTS: The grade groups were correctly estimated by the proposed approach without errors on training (70 ROIs) and validation (40 ROIs) data. CONCLUSIONS: Our results suggest that the proposed approach may support pathologists during the evaluation of grade groups for prostate cancer, thus mitigating intra- and interobserver variability.

CT features and quantitative analysis of subsolid nodule lung adenocarcinoma for pathological classification prediction.

BACKGROUND: The value of the CT features and quantitative analysis of lung subsolid nodules (SSNs) in the prediction of the pathological grading of lung adenocarcinoma is discussed. METHODS: Clinical data and CT images of 207 cases (216 lesions) with CT manifestations of an SSNs lung adenocarcinoma confirmed by surgery pathology were retrospectively analysed. The pathological results were divided into three groups, including atypical adenomatous hyperplasia (AAH)/adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC). Then, the quantitative and qualitative data of these nodules were compared and analysed. RESULTS: The mean size, maximum diameter, mean CT value and maximum CT value of the nodules were significantly different among the three groups of AAH/AIS, MIA and IAC and were different between the paired groups (AAH/AIS and MIA or MIA and IAC) (P < 0.05). The critical values of the above indicators between AAH/AIS and MIA were 10.05 mm, 11.16 mm, - 548.00 HU and - 419.74 HU. The critical values of the above indicators between MIA and IAC were 14.42 mm, 16.48 mm, - 364.59 HU and - 16.98 HU. The binary logistic regression analysis of the features with the statistical significance showed that the regression model between AAH/AIS and MIA is logit(p) = - 0.93 + 0.216X1 + 0.004X4. The regression model between MIA and IAC is logit(p) = - 1.242-1.428X5(1) - 1.458X6(1) + 1.146X7(1) + 0.272X2 + 0.005X3. The areas under the curve (AUC) obtained by plotting the receiver operating characteristic curve (ROC) using the regression probabilities of regression models I and II were 0.815 and 0.931. CONCLUSIONS: Preoperative prediction of pathological classification of CT image features has important guiding value for clinical management. Correct diagnosis results can effectively improve the patient survival rate. Through comprehensive analysis of the CT features and qualitative data of SSNs, the diagnostic accuracy of SSNs can be effectively improved. The logistic regression model established in this study can better predict the pathological classification of SSNs lung adenocarcinoma on CT, and the predictive value is significantly higher than the independent use of each quantitative factor.

Radiomics Based on Multimodal MRI for the Differential Diagnosis of Benign and Malignant Breast Lesions.

BACKGROUND: MRI-based radiomics has been used to diagnose breast lesions; however, little research combining quantitative pharmacokinetic parameters of dynamic contrast-enhanced MRI (DCE-MRI) and diffusion kurtosis imaging (DKI) exists. PURPOSE: To develop and validate a multimodal MRI-based radiomics model for the differential diagnosis of benign and malignant breast lesions and analyze the discriminative abilities of different MR sequences. STUDY TYPE: Retrospective. POPULATION: In all, 207 female patients with 207 histopathology-confirmed breast lesions (95 benign and 112 malignant) were included in the study. Then 159 patients were assigned to the training group, and 48 patients comprised the validation group. FIELD STRENGTH/SEQUENCE: T2 -weighted (T2 W), T1 -weighted (T1 W), diffusion-weighted MR imaging (b-values = 0, 500, 800, and 2000 seconds/mm2 ) and quantitative DCE-MRI were performed on a 3.0T MR scanner. ASSESSMENT: Radiomics features were extracted from T2 WI, T1 WI, DKI, apparent diffusion coefficient (ADC) maps, and DCE pharmacokinetic parameter maps in the training set. Models based on each sequence or combinations of sequences were built using a support vector machine (SVM) classifier and used to differentiate benign and malignant breast lesions in the validation set. STATISTICAL TESTS: Optimal feature selection was performed by Spearman's rank correlation coefficients and the least absolute shrinkage and selection operator algorithm (LASSO). Receiver operating characteristic (ROC) curves were used to assess the diagnostic performance of the radiomics models in the validation set. RESULTS: The area under the ROC curve (AUC) of the optimal radiomics model, including T2 WI, DKI, and quantitative DCE-MRI parameter maps was 0.921, with an accuracy of 0.833. The AUCs of the models based on T1 WI, T2 WI, ADC map, DKI, and DCE pharmacokinetic parameter maps were 0.730, 0.791, 0.770, 0.788, and 0.836, respectively. DATA CONCLUSION: The model based on radiomics features from T2 WI, DKI, and quantitative DCE pharmacokinetic parameter maps has a high discriminatory ability for benign and malignant breast lesions. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:596-607.

AF-SENet: Classification of Cancer in Cervical Tissue Pathological Images Based on Fusing Deep Convolution Features.

Cervical cancer is the fourth most common cancer in the world. Whole-slide images (WSIs) are an important standard for the diagnosis of cervical cancer. Missed diagnoses and misdiagnoses often occur due to the high similarity in pathological cervical images, the large number of readings, the long reading time, and the insufficient experience levels of pathologists. Existing models have insufficient feature extraction and representation capabilities, and they suffer from insufficient pathological classification. Therefore, this work first designs an image processing algorithm for data augmentation. Second, the deep convolutional features are extracted by fine-tuning pre-trained deep network models, including ResNet50 v2, DenseNet121, Inception v3, VGGNet19, and Inception-ResNet, and then local binary patterns and a histogram of the oriented gradient to extract traditional image features are used. Third, the features extracted by the fine-tuned models are serially fused according to the feature representation ability parameters and the accuracy of multiple experiments proposed in this paper, and spectral embedding is used for dimension reduction. Finally, the fused features are inputted into the Analysis of Variance-F value-Spectral Embedding Net (AF-SENet) for classification. There are four different pathological images of the dataset: normal, low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL), and cancer. The dataset is divided into a training set (90%) and a test set (10%). The serial fusion effect of the deep features extracted by Resnet50v2 and DenseNet121 () is the best, with average classification accuracy reaching 95.33%, which is 1.07% higher than ResNet50 v2 and 1.05% higher than DenseNet121. The recognition ability is significantly improved, especially in LSIL, reaching 90.89%, which is 2.88% higher than ResNet50 v2 and 2.1% higher than DenseNet121. Thus, this method significantly improves the accuracy and generalization ability of pathological cervical WSI recognition by fusing deep features.

Expert knowledge-infused deep learning for automatic lung nodule detection.

BACKGROUND: Computer aided detection (CADe) of pulmonary nodules from computed tomography (CT) is crucial for early diagnosis of lung cancer. Self-learned features obtained by training datasets via deep learning have facilitated CADe of the nodules. However, the complexity of CT lung images renders a challenge of extracting effective features by self-learning only. This condition is exacerbated for limited size of datasets. On the other hand, the engineered features have been widely studied. OBJECTIVE: We proposed a novel nodule CADe which aims to relieve the challenge by the use of available engineered features to prevent convolution neural networks (CNN) from overfitting under dataset limitation and reduce the running-time complexity of self-learning. METHODS: The CADe methodology infuses adequately the engineered features, particularly texture features, into the deep learning process. RESULTS: The methodology was validated on 208 patients with at least one juxta-pleural nodule from the public LIDC-IDRI database. Results demonstrated that the methodology achieves a sensitivity of 88% with 1.9 false positives per scan and a sensitivity of 94.01% with 4.01 false positives per scan. CONCLUSIONS: The methodology shows high performance compared with the state-of-the-art results, in terms of accuracy and efficiency, from both existing CNN-based approaches and engineered feature-based classifications.

Quantitative Image Feature Engine (QIFE): an Open-Source, Modular Engine for 3D Quantitative Feature Extraction from Volumetric Medical Images.

The aim of this study was to develop an open-source, modular, locally run or server-based system for 3D radiomics feature computation that can be used on any computer system and included in existing workflows for understanding associations and building predictive models between image features and clinical data, such as survival. The QIFE exploits various levels of parallelization for use on multiprocessor systems. It consists of a managing framework and four stages: input, pre-processing, feature computation, and output. Each stage contains one or more swappable components, allowing run-time customization. We benchmarked the engine using various levels of parallelization on a cohort of CT scans presenting 108 lung tumors. Two versions of the QIFE have been released: (1) the open-source MATLAB code posted to Github, (2) a compiled version loaded in a Docker container, posted to DockerHub, which can be easily deployed on any computer. The QIFE processed 108 objects (tumors) in 2:12 (h/mm) using 1 core, and 1:04 (h/mm) hours using four cores with object-level parallelization. We developed the Quantitative Image Feature Engine (QIFE), an open-source feature-extraction framework that focuses on modularity, standards, parallelism, provenance, and integration. Researchers can easily integrate it with their existing segmentation and imaging workflows by creating input and output components that implement their existing interfaces. Computational efficiency can be improved by parallelizing execution at the cost of memory usage. Different parallelization levels provide different trade-offs, and the optimal setting will depend on the size and composition of the dataset to be processed.