Automatic Analysis Research Articles

Automatic Quantitative Analysis of Internal Quantum Efficiency Measurements of GaAs Solar Cells Using Deep Learning.

A solar cell's internal quantum efficiency (IQE) measurement reveals critical information about the device's performance. This information can be obtained using a qualitative analysis of the shape of the curve, identifying and attributing current losses such as at the front and rear interfaces, and extracting key electrical and optical performance parameters. However, conventional methods to extract the performance parameters from IQE measurements are often time-consuming and require manual fitting approaches. While several methodologies exist to extract those parameters from silicon solar cells, there is a lack of accessible approaches for non-silicon cell technologies, like gallium arsenide cells, typically limiting the analysis to only the qualitative level. Therefore, this study proposes using a deep learning method to automatically predict multiple key parameters from IQE measurements of gallium arsenide cells. The proposed method is demonstrated to achieve a very high level of prediction accuracy across the entire range of parameter values and exhibits a high resilience for noisy measurements. By enhancing the quantitative analysis of IQE measurements, the method will unlock the full potential of quantum efficiency measurements as a powerful characterization tool for diverse solar cell technologies.

Clinical Value of Circ-PNPT1 on Adverse Pregnancy Outcomes of Patients with Gestational Diabetes Mellitus.

Several circular RNAs are associated with important pathophysiological characteristics of gestational diabetes mellitus (GDM). This study intended to measure the expression of circ-PNPT1 in sera of GDM patients and to expound on its values on pregnancy outcomes. Totally 104 GDM patients and 71 healthy controls were recruited. The expression pattern of serum circ-PNPT1 was measured by reverse transcription-quantitative polymerase chain reaction. The diagnostic efficacy of circ-PNPT1 and fasting blood glucose (FBG) on GDM was evaluated by receiver operating characteristic (ROC) analysis. Parameters of glycolipid metabolism were determined using automatic biochemical analyzers. The correlation between circ-PNPT1 and glycolipid metabolism parameters was analyzed using Pearson analysis. GDM patients were divided into a high expression group and a low expression group based on the median value of circ-PNPT1 expression. Curves of adverse neonatal outcomes were drawn by Log Rank analysis. GDM patients exhibited higher circ-PNPT1 expression than healthy controls. The area under the ROC curve of circ-PNPT1 diagnosing GDM was 0.9184 and the cut-off value was 1.435 (90.38% sensitivity, 85.92% specificity). Serum circ-PNPT1 expression was positively correlated with FBG, total cholesterol, and triglyceride in GDM patients. Neonates born to GDM patients with high circ- PNPT1 expression were prone to adverse outcomes. Circ-PNPT1 was highly-expressed in the sera of GDM patients. Circ-PNPT1 affected glycolipid metabolism and its expression had certain reference values on adverse pregnancy outcomes.

Bioindication Studies of Surface Water Supported by Automatic Image Analysis Using Deep Learning Neural Network – Cyclotella Case Study

Bioindication Studies of Surface Water Supported by Automatic Image Analysis Using Deep Learning Neural Network – Cyclotella Case Study

Commentary: Artificial Intelligence for Automatic Analysis of Shunt Treatment in Presurgery and Postsurgery Computed Tomography Brain Scans of Patients With Idiopathic Normal Pressure Hydrocephalus.

Commentary: Artificial Intelligence for Automatic Analysis of Shunt Treatment in Presurgery and Postsurgery Computed Tomography Brain Scans of Patients With Idiopathic Normal Pressure Hydrocephalus.

Stability, change, formation: Insights into the media's role in shaping attitudes toward green hydrogen in Germany

This study uses a multi-method design to investigate the media's role in shaping Germans' attitudes toward green hydrogen. It combines an automatized content analysis of 7649 German newspaper articles published between July 2021 and June 2024 and a three-wave panel survey of the German population conducted between June 2023 and June 2024 with an initial sample of 2701 participants. The findings show that the intensity of media reporting on hydrogen was low compared to other energy-related topics. Nevertheless, participants' assessments of relative topic presence are rather accurate (rho: 0.50–0.80). A considerable number of participants were unable to position themselves toward the potential and challenges of hydrogen (23%–35%). Overall, the results indicate that media and communication tend to stabilize or change existing attitudes rather than contribute to the formation or loss of attitudes, leading to implications for the communication of relevant stakeholders.

Weakly-Supervised Hair SEM Microscope Image Segmentation Using a Priori Structure Information

The analysis of microscopic images of hair has a wide range of applications in the domains of cosmetics, healthcare and forensics. The segmentation of the hair represents the initial step of automatic large-scale quantitative analysis of microscopic images of hair. It ensures that subsequent quantitative measurements are performed in an appropriate area corresponding to the hair, avoiding artifacts near the boundaries. This process can be time-consuming, tedious and susceptible to subjective errors when conducted by a human operator. Deep learning methods represent a promising solution; however, obtaining pixel-level accurate masks is a costly process. This paper presents a novel weakly supervised pipeline for the segmentation of hair SEM (Scanning Electron Microscope) microscopic images, which requires only simple image-level annotations for training. The proposed method incorporates the Radon transform, the Sobel operator and a novel Boundary Discrimination Module (BD-module) for the estimation of the presence of boundaries. The proposed pipeline was evaluated on a recently collected hair SEM dataset (429 images). Furthermore, it is benchmarked with methods including Unet and Segment Anything Model (SAM). The results demonstrated a mean Hausdorff Distance improvement of over 30% and a standard deviation improvement of over 50% in comparison with the Unet and SAM. Moreover, we proposed additional refinement modules to address boundary nonlinear cases and conducted Grad-CAM analysis to enhance the interpretability of the BD-module. Additionally, we proposed a novel quality estimation metric based on gradient map for self-quality assessment. The SEM hair dataset is accessible to the research community in an open-source format.

Unlocking AutoML: Enhancing Data with Deep Learning Algorithms for Medical Imaging

Deep learning algorithms have become increasingly popular over the years, having proved their efficiency in input-output functions for distinct types of data. This technology is particularly useful in medical imaging, where complex image structures often generate disagreements between medical staff. These technologies can streamline the diagnostic process by performing automatic image analysis, which results in more accurate and reproducible diagnoses. Additionally, these technologies can enhance content retrieval systems by automatically labeling the images based on the structures they possess. Despite the benefits, the mathematical complexity of deep learning algorithms and their training optimizations can be challenging. Automated machine learning provides a solution to this challenge by offering tools that automate the development and training of these algorithms. This makes it possible for users with limited programming experience to take advantage of these powerful technologies to quickly develop and prototype analysis algorithms for their specific needs. This paper presents a management platform for deep learning services on the cloud that provides a code-free experience through automated machine learning. The evaluation was done in one of the most demanding scenarios, where the service was integrated into a research pathology PACS to annotate mitotic cells in breast cancer tissue automatically. The annotations are processed by an open-source PACS archive and stored directly on the files, enhancing the image metadata and consequently content retrieval systems. The results of the developed algorithms were compared to the state-of-the-art to evaluate the competitiveness of the solution.

Fibronectin type III domain containing protein 5/irisin alleviated sepsis-induced acute kidney injury by abating ferroptosis through the adenosine 5'-monophosphate-activated protein kinase/nuclear factor erythroid-2-related factor 2 signaling pathway

Objective: Ferroptosis has been described in association with acute kidney injury (AKI)-induced sepsis. Fibronectin type III domain containing protein 5 (FNDC5)/irisin plays a crucial role in renal protection. The objective of this study was to investigate whether FNDC5/irisin is involved in AKI-induced sepsis by modulating ferroptosis, and the molecular mechanisms that may be involved. Material and Methods: A sepsis-induced AKI model was built in vivo and in vitro through lipopolysaccharide (LPS) intervention. FNDC5, adenosine 5'-monophosphate-activated protein kinase (AMPK), phospho-AMPK (p-AMPK), nuclear factor erythroid-2-related factor 2 (Nrf2), heme oxygenase-1 (HO-1), glutathione peroxidase 4 (GPX4), and acyl-CoA synthetase long-chain family member 4 (ACSL4) concentrations in cells and mouse kidney tissues were appraised by Western blot. Pro-inflammatory cytokines concentrations in cell supernatants and mouse kidney tissues were appraised by enzyme-linked immunosorbent assay. Fe2+ concentration in cells and mouse kidney tissue was appraised by kit. The apoptosis rate of cells and mouse kidney tissue was measured by flow cytometry. Automatic biochemical analyzer was to test serum creatinine (SCr) and blood urea nitrogen (BUN). The kidney tissue sections from each groups were observed by hematoxylin and eosin staining. Results: LPS abated FNDC5 concentration in human kidney-2 cells and mouse kidney tissue (P < 0.001). Overexpression of FNDC5 can abated proinflammatory cytokines concentrations in cells and mouse kidney tissue (P < 0.01). Meanwhile, overexpression of FNDC5 can boost GPX4 protein concentration, abate ACSL4 protein, and abate Fe2+ concentration in cells and mouse kidney tissues (P < 0.05). In addition, the overexpression of FNDC5 can reduce the rate of apoptosis (P < 0.01). In vivo experiments showed that FNDC5 overexpression reduced serum BUN and SCr concentrations and alleviated pathological damage in the mouse renal tissues (P < 0.05) and exhibited a certain renal protective effect. FNDC5 overexpression can boost p-AMPK/AMPK, Nrf2, and HO-1 protein concentrations (P < 0.01). Conclusion: FNDC5/irisin improves sepsis-induced acute renal injury by abating ferroptosis through the AMPK/Nrf2 signaling pathway.

Acoustic Tunnel Lining Void Detection: Modeling and Instrument System Development

The detachment of railway tunnel lining constitutes a grave danger to train operation safety and drastically curtails the tunnel’s service life. This study endeavors to efficiently detect the void defects in railway tunnel lining by creating a finite element model of tunnel lining structures. Utilizing this model, the study simulates the nonlinear acoustic wave propagation cloud maps for three representative tunnel lining structures: void-free, air void, and water void. This facilitates a thorough examination of the acoustic signal characteristics in the wavefield, time domain, and frequency domain. To satisfy the precision and efficiency demands of tunnel lining void detection, this study has devised and developed a portable acoustic detector that incorporates automatic analysis and processing capabilities and is furnished with a high-performance rare-earth magneto-strictive acoustic excitation device. This detection system can swiftly detect and assess typical void defects in tunnel lining. To further validate the effectiveness of this system, this study conducted lining defect detection in the Pingdao Railway Tunnel in the eastern Qinling Mountains. The test results show that the detection rate of this system for both air-filled and water-filled voids with a width of 1 m reached 100%, demonstrating its extremely high application value.

Infection Status, Etiological Analysis of Aeromonas Spp. in Foodborne Diarrhea Patients from 2019 to 2023 in Wenzhou.

The infection status and etiological analysis of Aeromonas spp. from foodborne diarrhea patients in Wenzhou were carried out to provide the etiological basis for healthy diet and clinical treatment. Aeromonas isolates (n = 41) collected from foodborne diarrhea patients were identified using the automatic bacteriologic analyzer and mass spectrometer. Species identification, multilocus sequence typing, prediction of virulence genes, and antimicrobial resistance genes were analyzed by the data of whole genome sequencing. The antibiotic resistance of these isolates was determined using miniaturization of the broth dilution susceptibility test. A total of 1829 stool samples of diarrhea patients were collected, and the detection rate of Aeromonas spp. was 2.24% (41/1829). Moreover, Aeromonas spp. are more easily detected in warmer months (from June to August), which were identified as follows: A. veronii (53.66%, 22/41), A. caviae (21.95%, 9/41), A. hydrophila (9.76%, 4/41), A. dhakensis (4.88%, 2/41), A. rivipollensis (4.88%, 2/41), A. enteropelogenes (2.44%, 1/41), and A. media (2.44%, 1/41). All strains can be divided into 38 sequence types, 31 of which were novel, suggesting that Aeromonas spp. had high genetic diversity, multiple clones, and various sources in diarrhea patients. High number of genetic diversity and resistance were found in the Aeromonas isolates. In addition, the category distribution of the virulence genes was significantly different among the seven species of Aeromonas. Aeromonas spp. had different degrees of resistance to antibiotics, and tetracycline was the most serious, with a resistance rate of 27%. What's more, for some antimicrobial classes in silico antimicrobial resistance gene detection was highly correlated with phenotypic antimicrobial resistance patterns with an overall sensitivity of 93.3% and a specificity of 66.7%. The findings from this research highlighted the importance for development of prevention and control strategies to reduce the risk of foodborne diarrhea caused by Aeromonas spp.

An automatized semantic analysis of two large-scale listening tests: A corpus-based study

This study examined the semantic features of the simulated mini-lectures in the listening sections of the International English Language Testing System (IELTS) and the Test of English as a Foreign Language (TOEFL) based on automatized semantic analysis to explore the content validity of the two tests. Two study corpora were utilized, the IELTS corpus with 56 mini-lectures (38,944 words) and the TOEFL corpus with 285 mini-lectures (207,296 words). The reference corpus comprised 59 lectures from the Michigan Corpus of Academic Spoken English (MICASE), totaling 571,354 words. The corpora were submitted to automatized semantic tagging using Wmatrix5. Three comparisons were conducted: IELTS versus TOEFL, IELTS versus MICASE lectures, and TOEFL versus MICASE lectures. The results suggest that IELTS and TOEFL mini-lectures shared 78% and 64% of the same semantic features as MICASE, respectively, supporting their relative content validity. Nevertheless, specific semantic categories, such as politics, war, and intimate and sexual relationships, were notably absent from the test corpora, even though they appeared in the academic lecture corpus. In addition, causal connectors are frequently used in both tests, while the mini-lectures of IELTS listening tests cover fewer academic discourse fields than TOEFL mini-lectures. Implications for content validity are discussed.

Neural network embedding of functional microconnectome

Abstract Our brains operate as a complex network of interconnected neurons. To gain a deeper understanding of this network architecture, it is essential to extract simple rules from its intricate structure. This study aimed to compress and simplify the architecture, with a particular focus on interpreting patterns of functional connectivity in 2.5 hr of electrical activity from a vast number of neurons in acutely sliced mouse brains. Here, we combined two distinct methods together: automatic compression and network analysis. Firstly, for automatic compression, we trained an artificial neural network named NNE (neural network embedding). This allowed us to reduce the connectivity to features, be represented only by 13% of the original neuron count. Secondly, to decipher the topology, we concentrated on the variability among the compressed features and compared them with 15 distinct network metrics. Specifically, we introduced new metrics that had not previously existed, termed as indirect-adjacent degree and neighboring hub ratio. Our results conclusively demonstrated that these new metrics could better explain approximately 40%–45% of the features. This finding highlighted the critical role of NNE in facilitating the development of innovative metrics, because some of the features extracted by NNE were not captured by the currently existed network metrics.

Immunomodulatory effect of compound Xuelian capsule on experimental autoimmune myasthenia gravis in mice

IntroductionThis study aimed to explore the immunomodulatory effects of Xuelian capsules in experimental autoimmune myasthenia gravis (EAMG) mice. MethodsFemale C57BL/6 mice were immunized with AChR-α subunit R97–116 peptide to induce EAMG. Mice were randomized into five groups: EAMG model (n = 8), prednisone (n = 7), and low-, middle-, and high-dose Xuelian capsule groups (n = 7, 7, 8). Hematological parameters and NLR were assessed using an automatic hematology analyzer. Serum AchR-Ab, IFN-γ, IL-4, IL-10 levels were measured by ELISA. Thymus pathology was evaluated by HE staining. Spleen mRNA levels of TGF-β, IL-17, IL-6, IFN-γ were quantified by qPCR, and spleen cell populations by flow cytometry. ResultsXuelian capsules significantly reduced the ratio of NLR in peripheral blood, the concentration of AchR-Ab, IFN-γ, and IL-4 in serum, the expression of TGF-β, IL-17, IL-6, and IFN-γ in the spleen, and the proportion of CD4+/IL-4+ and CD4+/IFN-γ+ cells in the spleen, while increasing the concentration of IL-10 in the serum and the proportion of CD4+/Treg cells in the spleen. It also improved pathological changes in the thymus tissue of mice. DiscussionOne of the possible mechanisms of Xuelian capsules in the treatment of myasthenia gravis is to improve the pathological structure of the thymus of EAMG mice by inhibiting inflammatory factors and improving immunity, which then affects the occurrence and development of inflammation to achieve anti-inflammation, thus improving myasthenia gravis (MG).

Novel analysis tool for the distance of gold dimers controlled by the DNA strand length on the DNA origami.

Metallic nanoparticle dimers have been used to enhance the excitation rate of single-quantum emitters. The interparticle distance (d) of the dimers has a crucial influence on the signal enhancement. Therefore, precise control of d is desired for optimal performance. However, statistical analysis of d has been often restricted to a small number of dimers due to the lack of reliable automatic software tools. For this reason, we developed a novel analysis tool for automatic dimer analysis. Our approach combines particle detection by circle Hough transformation (CHT) with custom classification routines optimised for distinct types of particles. We applied our tool to scanning electron microscopy (SEM) images and achieved great agreement in dimer detection, reaching an agreement of around 97% between automatic analysis and manual inspection for more than 3000 metallic nanoparticle dimers on DNA origami controlled by a combination of multiple DNA strands. Our study revealed the effects of the strand length (L) on the distribution of d. Based on geometric consideration, we expected a strong correlation between L and the standard deviation (σ) of d. We could verify this correlation by characterising four dimer designs with different L while analysing more than 1000 dimers per specimen.

Artificial Intelligence Applied to Support Agronomic Decisions for the Automatic Aerial Analysis Images Captured by UAV: A Systematic Review

Integrating advanced technologies such as artificial intelligence (AI) with traditional agricultural practices has changed how activities are developed in agriculture, with the aim of automating manual processes and improving the efficiency and quality of farming decisions. With the advent of deep learning models such as convolutional neural network (CNN) and You Only Look Once (YOLO), many studies have emerged given the need to develop solutions to problems and take advantage of all the potential that this technology has to offer. This systematic literature review aims to present an in-depth investigation of the application of AI in supporting the management of weeds, plant nutrition, water, pests, and diseases. This systematic review was conducted using the PRISMA methodology and guidelines. Data from different papers indicated that the main research interests comprise five groups: (a) type of agronomic problems; (b) type of sensor; (c) dataset treatment; (d) evaluation metrics and quantification; and (e) AI technique. The inclusion (I) and exclusion (E) criteria adopted in this study included: (I1) articles that obtained AI techniques for agricultural analysis; (I2) complete articles written in English; (I3) articles from specialized scientific journals; (E1) articles that did not describe the type of agrarian analysis used; (E2) articles that did not specify the AI technique used and that were incomplete or abstract; (E3) articles that did not present substantial experimental results. The articles were searched on the official pages of the main scientific bases: ACM, IEEE, ScienceDirect, MDPI, and Web of Science. The papers were categorized and grouped to show the main contributions of the literature to support agricultural decisions using AI. This study found that AI methods perform better in supporting weed detection, classification of plant diseases, and estimation of agricultural yield in crops when using images captured by Unmanned Aerial Vehicles (UAVs). Furthermore, CNN and YOLO, as well as their variations, present the best results for all groups presented. This review also points out the limitations and potential challenges when working with deep machine learning models, aiming to contribute to knowledge systematization and to benefit researchers and professionals regarding AI applications in mitigating agronomic problems.

A Similar Feature Point Matching Method for Aerial Electric Power Tower Images Based on a One-Ring Neighborhood of Vertices

Due to the susceptibility of images to various factors such as scale changes, imaging conditions, and image noise, traditional feature point matching methods are difficult to achieve ideal matching accuracy, leading to many challenges in the automatic analysis and processing of power tower images. To improve the matching accuracy of similar feature points in aerial images of electric power tower, this study proposes a method for matching similar feature points in aerial images of electric power tower based on a vertex ring neighborhood, addressing the problem of large matching errors caused by factors such as scale and imaging conditions. This method first adopts the feature point extraction technique of electric power tower image based on decomposition and filtering, and constructs the Laplacian pyramid of the original image. Subsequently, filter banks are used to decompose the pyramid image in different directions, extract local extremum points as candidate feature point sets, and merge them to obtain the final feature point set. On this basis, taking into account the spatial relationships and geometric characteristics around the feature points, a texture mapping algorithm based on vertex ring neighborhood and patch classification is applied to construct a vertex ring neighborhood structure and extract geometrically representative electric power tower features. Finally, by using a texture feature point matching method based on the principle of similarity, the similarity between the real-time image and the reference image features is calculated for matching, and combined with the Babbitt coefficient to remove mismatched feature points, accurate matching of similar feature points in aerial electric power tower images is achieved. The experimental results show that this method has a mean square error of less than 0.1Pixel in matching similar texture feature points of aerial power tower images under various working conditions, significantly improving the matching accuracy and providing an effective tool for automatic analysis and processing of power tower images.

Uricase-Expressing Engineered Macrophages Alleviate Murine Hyperuricemia.

Background: Uricase, or urate oxidase (Uox) is a key enzyme in uric acid (UA) metabolism and has been applied in clinical treatment of human hyperuricemia (HUA). However, the current clinically applied uricases, despite their potent urate-lowering capacity, tend to form anti-drug antibodies because of their immunogenicity, leading to increased risk of anaphylaxis, faster drug clearance and reduced or even complete loss of therapeutic effect, limiting their clinical application. In this study, we constructed engineered macrophages that stably expressed uricase, which might serve as a promising alternative to the direct injection of uricases. Materials and Methods: Engineered macrophages RAW264.7 cells were injected intravenously to treat hyperuricemic KM mice. Serum uric acid and bio-indicators for renal and hepatic functions were detected by an automatic biochemical analyzer; inflammatory cytokines were determined by ELISA; the livers and kidneys of the mice were sectioned for histological examination. Results: The uricase-expressing macrophages reduced UA levels from 300 ± 1.5 μmol/L to 101 ± 8.3 μmol/L in vitro. And in an HUA mouse model established by gavage with yeast extract, intravenous injection of the engineered macrophages could reduce the serum uric acid (sUA) of mice to normal level on the 14th day of modeling, with a decrease of 48.6%, and the urate-lowering effect was comparable to that of the first-line clinical drug allopurinol. In terms of safety, engineered macrophages did not cause liver or kidney dysfunction in mice, nor did they induce systemic immune response. Conclusions: Using macrophages as a chassis to deliver uricase might be a new, safe and effective strategy for the treatment and control of hyperuricemia.

RF sensing enabled tracking of human facial expressions using machine learning algorithms

Automatic analysis of facial expressions has emerged as a prominent research area in the past decade. Facial expressions serve as crucial indicators for understanding human behavior, enabling the identification and assessment of positive and negative emotions. Moreover, facial expressions provide insights into various aspects of mental activities, social connections, and physiological information. Currently, most facial expression detection systems rely on cameras and wearable devices. However, these methods have drawbacks, including privacy concerns, issues with poor lighting and line of sight blockage, difficulties in training with longer video sequences, computational complexities, and disruptions to daily routines. To address these challenges, this study proposes a novel and privacy-preserving human behavior recognition system that utilizes Frequency Modulated Continuous Wave (FMCW) radar combined with Machine Learning (ML) techniques for classifying facial expressions. Specifically, the study focuses on five common facial expressions: Happy, Sad, Fear, Surprise, and Neutral. The recorded data is obtained in the form of a Micro-Doppler signal, and state-of-the-art ML models such as Super Learner, Linear Discriminant Analysis, Random Forest, K-Nearest Neighbor, Long Short-Term Memory, and Logistic Regression are employed to extract relevant features. These extracted features from the radar data are then fed into ML models for classification. The results show a highly promising classification accuracy of 91%. The future applications of the proposed work will lead to advancements in technology, healthcare, security, and communication, thereby improving overall human well-being and societal functioning.

Automatic segmentation-based multi-modal radiomics analysis of US and MRI for predicting disease-free survival of breast cancer: a multicenter study

BackgroundSeveral studies have confirmed the potential value of applying radiomics to predict prognosis of breast cancer. However, the tumor segmentation in these studies depended on delineation or annotation of breast cancer by radiologist, which is often laborious, tedious, and vulnerable to inter- and intra-observer variability. Automatic segmentation is expected to overcome this difficulty. Herein, we aim to investigate the value of automatic segmentation-based multi-modal radiomics signature and magnetic resonance imaging (MRI) features in predicting disease-free survival (DFS) of patients diagnosed with invasive breast cancer.MethodsThis retrospective multicenter study included a total of 643 female patients with invasive breast cancer who underwent preoperative ultrasound (US) and MRI for prognostic analysis. Data (n = 480) from center 1 were divided into training and internal testing sets, while data (n = 163) from centers 2 and 3 were analyzed as the external testing set. We developed automatic segmentation frameworks for tumor segmentation by deep learning. Then, Least absolute shrinkage and selection operator Cox regression was used to select features to construct radiomics signature, and corresponding radiomics score (Rad-score) was calculated. Finally, six models for predicting DFS were constructed by using Cox regression and assessed in terms of discrimination, calibration, and clinical usefulness.ResultsThe multi-modal radiomics signature combining intra- and peri-tumoral radiomics signatures of US and MRI achieved a higher C-index in the internal (0.734) and external (0.708) testing sets than most other radiomics signatures in predicting DFS, and successfully stratified patients into low- and high-risk groups. The multi-modal clinical imaging model combining the multi-modal Rad-score and clinical traditional MRI model-score resulted in a higher C-index (0.795) than other models in the external testing set, and it had a better calibration and higher clinical benefit.ConclusionsThis study demonstrates that the multi-modal radiomics signature derived from automatic segmentations of US and MRI is a promising risk stratification biomarker for breast cancer, and highlights that the appropriate combination of multi-modal radiomics signature, clinical characteristics, and MRI feature can improve performance of individualized DFS prediction, which might assist in guiding decision-making related to breast cancer.

An open annotated dataset and baseline machine learning model for segmentation of vertebrae with metastatic bone lesions from CT.

Automatic analysis of pathologic vertebrae from computed tomography (CT) scans could significantly improve the diagnostic management of patients with metastatic spine disease. We provide the first publicly available annotated imaging dataset of cancerous CT spines to help develop artificial intelligence frameworks for automatic vertebrae segmentation and classification. This collection contains a dataset of 55 CT scans collected on patients with various types of primary cancers at two different institutions. In addition to raw images, data include manual segmentations and contours, vertebral level labeling, vertebral lesion-type classifications, and patient demographic details. Our automated segmentation model uses nnU-Net, a freely available open-source framework for deep learning in healthcare imaging, and is made publicly available. This data will facilitate the development and validation of models for predicting the mechanical response to loading and the resulting risk of fractures and spinal deformity.