The Usefulness of the Breast Density Assessment Application Used by Breast Radiologists.

Breast density determined by breast radiologists and also automatically estimated by applications has been widely investigated. However, no study has yet clarified whether the use of these applications by breast radiologists improves reading efficacy. Therefore, this study aimed to assess the usefulness of applications when used by breast radiologists. A Breast Density Assessment application (App) developed by Konica Minolta, Inc. (Tokyo, Japan) was used. Independent and sequential tests were conducted to assess the usefulness of the concurrent- and second-look modes. Fifty and 100 cases were evaluated using sequential and independent tests, respectively. Each dataset was configured based on the evaluation by an expert breast radiologist who developed the Japanese guidelines for breast density. Nine breast radiologists evaluated the mammary gland content ratio and breast density; the inter-observer and expert-to-observer variability were calculated. The time required to complete the experiments was also recorded. The inter-observer variability was significant with the App, as revealed by the independent test. The use of the App significantly improved the agreement between the responses of the observers for the mammary gland content ratio and those of the expert by 6.6% and led to a reduction of 186.9 seconds in the average time required by the observers to evaluate 100 cases. However, the results of the sequential test did not suggest the effectiveness of the App. These findings suggest that the concurrent use of the App improves reading efficiency.

Extracting fetal heart signals from Doppler using semi-supervised convolutional neural networks.

Cardiotocography (CTG) measurements are critical for assessing fetal wellbeing during monitoring, and accurate assessment requires well-traceable CTG signals. The current FHR calculation algorithm, based on autocorrelation to Doppler ultrasound (DUS) signals, often results in periods of loss owing to its inability to differentiate signals. We hypothesized that classifying DUS signals by type could be a solution and proposed that an artificial intelligence (AI)-based approach could be used for classification. However, limited studies have incorporated the use of AI for DUS signals because of the limited data availability. Therefore, this study focused on evaluating the effectiveness of semi-supervised learning in enhancing classification accuracy, even in limited datasets, for DUS signals. Data comprising fetal heartbeat, artifacts, and two other categories were created from non-stress tests and labor DUS signals. With labeled and unlabeled data totaling 9,600 and 48,000 data points, respectively, the semi-supervised learning model consistently outperformed the supervised learning model, achieving an average classification accuracy of 80.9%. The preliminary findings indicate that applying semi-supervised learning to the development of AI models using DUS signals can achieve high generalization accuracy and reduce the effort. This approach may enhance the quality of fetal monitoring.

The Approach to Sensing the True Fetal Heart Rate for CTG Monitoring: An Evaluation of Effectiveness of Deep Learning with Doppler Ultrasound Signals.

Cardiotocography (CTG) is widely used to assess fetal well-being. CTG is typically obtained using ultrasound and autocorrelation methods, which extract periodicity from the signal to calculate the heart rate. However, during labor, maternal vessel pulsations can be measured, resulting in the output of the maternal heart rate (MHR). Since the autocorrelation output is displayed as fetal heart rate (FHR), there is a risk that obstetricians may mistakenly evaluate the fetal condition based on MHR, potentially overlooking the necessity for medical intervention. This study proposes a method that utilizes Doppler ultrasound (DUS) signals and artificial intelligence (AI) to determine whether the heart rate obtained by autocorrelation is of fetal origin. We developed a system to simultaneously record DUS signals and CTG and obtained data from 425 cases. The midwife annotated the DUS signals by auditory differentiation, providing data for AI, which included 30,160 data points from the fetal heart and 2160 data points from the maternal vessel. Comparing the classification accuracy of the AI model and a simple mathematical method, the AI model achieved the best performance, with an area under the curve (AUC) of 0.98. Integrating this system into fetal monitoring could provide a new indicator for evaluating CTG quality.

Development of a Subtraction Processing Technology for Assistance in the Comparative Interpretation of Mammograms.

A comparative interpretation of mammograms has become increasingly important, and it is crucial to develop subtraction processing and registration methods for mammograms. However, nonrigid image registration has seldom been applied to subjects constructed with soft tissue only, such as mammograms. We examined whether subtraction processing for the comparative interpretation of mammograms can be performed using nonrigid image registration. As a preliminary study, we evaluated the results of subtraction processing by applying nonrigid image registration to normal mammograms, assuming a comparative interpretation between the left and right breasts. Mediolateral-oblique-view mammograms were taken from noncancer patients and divided into 1000 cases for training, 100 cases for validation, and 500 cases for testing. Nonrigid image registration was applied to align the horizontally flipped left-breast mammogram with the right one. We compared the sum of absolute differences (SAD) of the difference of bilateral images (Difference Image) with and without the application of nonrigid image registration. Statistically, the average SAD was significantly lower with the application of nonrigid image registration than without it (without: 0.0692; with: 0.0549 (p < 0.001)). In four subgroups using the breast area, breast density, compressed breast thickness, and Difference Image without nonrigid image registration, the average SAD of the Difference Image was also significantly lower with nonrigid image registration than without it (p < 0.001). Nonrigid image registration was found to be sufficiently useful in aligning bilateral mammograms, and it is expected to be an important tool in the development of a support system for the comparative interpretation of mammograms.

Spirometry test values can be estimated from a single chest radiograph.

Introduction: Physical measurements of expiratory flow volume and speed can be obtained using spirometry. These measurements have been used for the diagnosis and risk assessment of chronic obstructive pulmonary disease and play a crucial role in delivering early care. However, spirometry is not performed frequently in routine clinical practice, thereby hindering the early detection of pulmonary function impairment. Chest radiographs (CXRs), though acquired frequently, are not used to measure pulmonary functional information. This study aimed to evaluate whether spirometry parameters can be estimated accurately from single frontal CXR without image findings using deep learning. Methods: Forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), and FEV1/FVC as spirometry measurements as well as the corresponding chest radiographs of 11,837 participants were used in this study. The data were randomly allocated to the training, validation, and evaluation datasets at an 8:1:1 ratio. A deep learning network was pretrained using ImageNet. The input and output information were CXRs and spirometry test values, respectively. The training and evaluation of the deep learning network were performed separately for each parameter. The mean absolute error rate (MAPE) and Pearson's correlation coefficient (r) were used as the evaluation indices. Results: The MAPEs between the spirometry measurements and AI estimates for FVC, FEV1 and FEV1/FVC were 7.59% (r = 0.910), 9.06% (r = 0.879) and 5.21% (r = 0.522), respectively. A strong positive correlation was observed between the measured and predicted indices of FVC and FEV1. The average accuracy of >90% was obtained in each estimation of spirometry indices. Bland-Altman analysis revealed good agreement between the estimated and measured values for FVC and FEV1. Discussion: Frontal CXRs contain information related to pulmonary function, and AI estimation performed using frontal CXRs without image findings could accurately estimate spirometry values. The network proposed for estimating pulmonary function in this study could serve as a recommendation for performing spirometry or as an alternative method, suggesting its utility.

Identifying factors that indicate the possibility of non-visible cases on mammograms using mammary gland content ratio estimated by artificial intelligence.

Background: Mammography is the modality of choice for breast cancer screening. However, some cases of breast cancer have been diagnosed through ultrasonography alone with no or benign findings on mammography (hereby referred to as non-visibles). Therefore, this study aimed to identify factors that indicate the possibility of non-visibles based on the mammary gland content ratio estimated using artificial intelligence (AI) by patient age and compressed breast thickness (CBT). Methods: We used AI previously developed by us to estimate the mammary gland content ratio and quantitatively analyze 26,232 controls and 150 non-visibles. First, we evaluated divergence trends between controls and non-visibles based on the average estimated mammary gland content ratio to ensure the importance of analysis by age and CBT. Next, we evaluated the possibility that mammary gland content ratio ≥50% groups affect the divergence between controls and non-visibles to specifically identify factors that indicate the possibility of non-visibles. The images were classified into two groups for the estimated mammary gland content ratios with a threshold of 50%, and logistic regression analysis was performed between controls and non-visibles. Results: The average estimated mammary gland content ratio was significantly higher in non-visibles than in controls when the overall sample, the patient age was ≥40 years and the CBT was ≥40 mm (p < 0.05). The differences in the average estimated mammary gland content ratios in the controls and non-visibles for the overall sample was 7.54%, the differences in patients aged 40-49, 50-59, and ≥60 years were 6.20%, 7.48%, and 4.78%, respectively, and the differences in those with a CBT of 40-49, 50-59, and ≥60 mm were 6.67%, 9.71%, and 16.13%, respectively. In evaluating mammary gland content ratio ≥50% groups, we also found positive correlations for non-visibles when controls were used as the baseline for the overall sample, in patients aged 40-59 years, and in those with a CBT ≥40 mm (p < 0.05). The corresponding odds ratios were ≥2.20, with a maximum value of 4.36. Conclusion: The study findings highlight an estimated mammary gland content ratio of ≥50% in patients aged 40-59 years or in those with ≥40 mm CBT could be indicative factors for non-visibles.

Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence.

Recently, breast types were categorized into four types based on the Breast Imaging Reporting and Data System (BI-RADS) atlas, and evaluating them is vital in clinical practice. A Japanese guideline, called breast composition, was developed for the breast types based on BI-RADS. The guideline is characterized using a continuous value called the mammary gland content ratio calculated to determine the breast composition, therefore allowing a more objective and visual evaluation. Although a discriminative deep convolutional neural network (DCNN) has been developed conventionally to classify the breast composition, it could encounter two-step errors or more. Hence, we propose an alternative regression DCNN based on mammary gland content ratio. We used 1476 images, evaluated by an expert physician. Our regression DCNN contained four convolution layers and three fully connected layers. Consequently, we obtained a high correlation of 0.93 (p < 0.01). Furthermore, to scrutinize the effectiveness of the regression DCNN, we categorized breast composition using the estimated ratio obtained by the regression DCNN. The agreement rates are high at 84.8%, suggesting that the breast composition can be calculated using regression DCNN with high accuracy. Moreover, the occurrence of two-step errors or more is unlikely, and the proposed method can intuitively understand the estimated results.

[Selection of Radiomic Features for the Classification of Triple-negative Breast Cancer Based on Radiogenomics].

Subtype classification of breast cancer by analyzing the gene expression profile of cancer cells is becoming a standard procedure. Breast cancer subtype classification is more useful than the conventional method because the characteristics of subtype classification is directly connected with the treatment method. However, genetic testing is invasive, and a part of cancer cells may not represent the overall nature of the cancer. In the computer-aided diagnosis (CAD) scheme for differentiation of triple-negative breast cancer (TNBC) by estimating the genetic properties of cancer based on Radiogenomics, principal component analysis (PCA) and least absolute shrinkage and selection operator (Lasso) were used for reducing the dimension of radiomic features, and we compared usefulness of both. We collected 81 magnetic resonance (MR) images, which included 30 TNBC and 51 others, from the public database. From the MR slice images, we selected the slice containing the largest area of the cancer and manually marked the cancer region. We subsequently calculated 294 radiomic features in the cancer region, and reduced the dimension of radiomic features. Finally, linear discriminant analysis, with the dimensionally compressed 10 image features, was used for distinguishing between TNBC and others. Area under the curve (AUC) was 0.60 when we used PCA, whereas AUC was 0.70 when we used Lasso (p=0.0058). Therefore, Lasso is useful for the determination of radiomic features in Radiogenomics.

[Quantitation of Cerebral Atrophy due to Normal Aging: Principal Component Analysis with MR Images in Patients' Age Groups].

To evaluate the degree of cerebral atrophy, quantification methods of a difference from a standard normal brain are often used in clinical practice. However, these methods may not evaluate the cerebral atrophy accurately, because they do not take into account any cerebral atrophies due to normal aging. The purpose of this study is to develop a model for taking into account the cerebral atrophy due to normal aging. We obtained 60 normal magnetic resonance (MR) images from the Alzheimer's disease neuroimaging initiative database. These data included 20 images of each age group of 60's, 70's, and 80's, respectively. For anatomical standardization of the images, we used the statistical parametric mapping software and employed a linear grayscale transformation. The principal component (PC) analysis with voxel values of 60 normal MR images was subsequently performed to calculate eigenvectors and PC scores. All cases were projected onto the eigenspace formed by 2nd and 5th PC scores. The experimental result showed separated distributions corresponding to the age groups. In addition, the sites of cerebral atrophy could be recognized by displaying eigenimages. Our proposed method would be useful for the accurate evaluation of cerebral atrophy caused by Alzheimer's disease.

Computer-aided diagnosis with radiogenomics: analysis of the relationship between genotype and morphological changes of the brain magnetic resonance images.

In the post-genome era, a novel research field, 'radiomics' has been developed to offer a new viewpoint for the use of genotypes in radiology and medicine research which have traditionally focused on the analysis of imaging phenotypes. The present study analyzed brain morphological changes related to the individual's genotype. Our data consisted of magnetic resonance (MR) images of patients with mild cognitive impairment (MCI) and Alzheimer's disease (AD), as well as their apolipoprotein E (APOE) genotypes. First, statistical parametric mapping (SPM) 12 was used for three-dimensional anatomical standardization of the brain MR images. A total of 30 normal images were used to create a standard normal brain image. Z-score maps were generated to identify the differences between an abnormal image and the standard normal brain. Our experimental results revealed that cerebral atrophies, depending on genotypes, can occur in different locations and that morphological changes may differ between MCI and AD. Using a classifier to characterize cerebral atrophies related to an individual's genotype, we developed a computer-aided diagnosis (CAD) scheme to identify the disease. For the early detection of cerebral diseases, a screening system using MR images, called Brain Check-up, is widely performed in Japan. Therefore, our proposed CAD scheme would be used in Brain Check-up.