BMC Med Imaging. 2024 Nov 18;24(1):314. doi: 10.1186/s12880-024-01486-z.

ABSTRACT

In recent years, the incidence of nodular thyroid diseases has been increasing annually. Ultrasonography has become a routine diagnostic tool for thyroid nodules due to its high real-time capabilities and low invasiveness. However, thyroid images obtained from current ultrasound tests often have low resolution and are plagued by significant noise interference. Regional differences in medical conditions and varying levels of physician experience can impact the accuracy and efficiency of diagnostic results. With the advancement of deep learning technology, deep learning models are used to identify whether a nodule in a thyroid ultrasound image is benign or malignant. This helps to close the gap between doctors' experience and equipment differences, improving the accuracy of the initial diagnosis of thyroid nodules. To cope with the problem that thyroid ultrasound images contain complex background and noise as well as poorly defined local features. in this paper, we first construct an improved ResNet50 classification model that uses a two-branch input and incorporates a global attention lightening module. This model is used to improve the accuracy of benign and malignant nodule classification in thyroid ultrasound images and to reduce the computational effort due to the two-branch structure.We constructed a U-net segmentation model incorporating our proposed ACR module, which uses hollow convolution with different dilation rates to capture multi-scale contextual information for feature extraction of nodules in thyroid ultrasound images and uses the results of the segmentation task as an auxiliary branch for the classification task to guide the classification model to focus on the lesion region more efficiently in the case of weak local features. The classification model is guided to focus on the lesion region more efficiently, and the classification and segmentation sub-networks are respectively improved specifically for this study, which is used to improve the accuracy of classifying the benign and malignant nature of the nodules in thyroid ultrasound images. The experimental results show that the four evaluation metrics of accuracy, precision, recall, and f1 of the improved model are 96.01%, 93.3%, 98.8%, and 96.0%, respectively. The improvements were 5.7%, 1.6%, 13.1%, and 7.4%, respectively, compared with the baseline classification model.

PMID:39558260 | DOI:10.1186/s12880-024-01486-z

BMC Med Imaging. 2024 Nov 18;24(1):313. doi: 10.1186/s12880-024-01473-4.

ABSTRACT

BACKGROUND: Esophageal fistula (EF), a rare and potentially fatal complication, can be better managed with predictive models for personalized treatment plans in esophageal cancers. We aim to develop a clinical-deep learning radiomics model for effectively predicting the occurrence of EF.

METHODS: The study involved esophageal cancer patients undergoing radiotherapy or chemoradiotherapy. Arterial phase enhanced CT images were used to extract handcrafted and deep learning radiomic features. Along with clinical information, a 3-step feature selection method (statistical tests, Least Absolute Shrinkage and Selection Operator, and Recursive Feature Elimination) was used to identify five feature sets in training cohort for constructing random forest EF prediction models. Model performance was compared and validated in both retrospective and prospective test cohorts.

RESULTS: One hundred seventy five patients (122 in training and 53 in test cohort)were retrospectively collected from April 2018 to June 2022. An additional 27 patients were enrolled as a prospective test cohort from June 2022 to December 2023. Post-selection in the training cohort, five feature sets were used for model construction: clinical, handcrafted radiomic, deep learning radiomic, clinical-handcrafted radiomic, and clinical-deep learning radiomic. The clinical-deep learning radiomic model excelled with AUC of 0.89 (95% Confidence Interval: 0.83-0.95) in the training cohort, 0.81 (0.65-0.94) in the test cohort, and 0.85 (0.71-0.97) in the prospective test cohort. Brier-score and calibration curve analyses validated its predictive ability.

CONCLUSIONS: The clinical-deep learning radiomic model can effectively predict EF in patients with advanced esophageal cancer undergoing radiotherapy or chemoradiotherapy.

PMID:39558242 | DOI:10.1186/s12880-024-01473-4

BMC Med Imaging. 2024 Nov 18;24(1):312. doi: 10.1186/s12880-024-01469-0.

ABSTRACT

BACKGROUND AND PURPOSE: Tumor bed (TB) is the residual cavity of resected tumor after surgery. Delineating TB from CT is crucial in generating clinical target volume for radiotherapy. Due to multiple surgical effects and low image contrast, segmenting TB from soft tissue is challenging. In clinical practice, titanium clips were used as marks to guide the searching of TB. However, this information is limited and may cause large error. To provide more prior location information, the tumor regions on both pre-operative and post-operative CTs are both used by the deep learning model in segmenting TB from surrounding tissues.

MATERIALS AND METHODS: For breast cancer patient after surgery and going to be treated by radiotherapy, it is important to delineate the target volume for treatment planning. In clinical practice, the target volume is usually generated from TB by adding a certain margin. Therefore, it is crucial to identify TB from soft tissue. To facilitate this process, a deep learning model is developed to segment TB from CT with the guidance of prior tumor location. Initially, the tumor contour on the pre-operative CT is delineated by physician for surgical planning purpose. Then this contour is transformed to the post-operative CT via the deformable image registration between paired pre-operative and post-operative CTs. The original and transformed tumor regions are both used as inputs for predicting the possible region of TB by the deep-learning model.

RESULTS: Compared to the one without prior tumor contour information, the dice similarity coefficient of the deep-learning model with the prior tumor contour information is improved significantly (0.812 vs. 0.520, P = 0.001). Compared to the traditional gray-level thresholding method, the dice similarity coefficient of the deep-learning model with the prior tumor contour information is improved significantly (0.812 vs.0.633, P = 0.0005).

CONCLUSIONS: The prior tumor contours on both pre-operative and post-operative CTs provide valuable information in searching for the precise location of TB on post-operative CT. The proposed method provided a feasible way to assist auto-segmentation of TB in treatment planning of radiotherapy after breast-conserving surgery.

PMID:39558240 | DOI:10.1186/s12880-024-01469-0

Nucleic Acids Res. 2024 Nov 21:gkae1027. doi: 10.1093/nar/gkae1027. Online ahead of print.

ABSTRACT

Response to spatiotemporal variation in selection gradients resulted in signatures of polygenic adaptation in human genomes. We introduce RAISING, a two-stage deep learning framework that optimizes neural network architecture through hyperparameter tuning before performing feature selection and prediction tasks. We tested RAISING on published and newly designed simulations that incorporate the complex interplay between demographic history and selection gradients. RAISING outperformed Phylogenetic Generalized Least Squares (PGLS), ridge regression and DeepGenomeScan, with significantly higher true positive rates (TPR) in detecting genetic adaptation. It reduced computational time by 60-fold and increased TPR by up to 28% compared to DeepGenomeScan on published data. In more complex demographic simulations, RAISING showed lower false discoveries and significantly higher TPR, up to 17-fold, compared to other methods. RAISING demonstrated robustness with least sensitivity to demographic history, selection gradient and their interactions. We developed a sliding window method for genome-wide implementation of RAISING to overcome the computational challenges of high-dimensional genomic data. Applied to African, European, South Asian and East Asian populations, we identified multiple genomic regions undergoing polygenic selection. Notably, ∼70% of the regions identified in Africans are unique, with broad patterns distinguishing them from non-Africans, corroborating the Out of Africa dispersal model.

PMID:39558170 | DOI:10.1093/nar/gkae1027

Sci Rep. 2024 Nov 18;14(1):28428. doi: 10.1038/s41598-024-80132-8.

ABSTRACT

The current investigation proposes a novel hybrid methodology for the diagnosis of the foot fractures. The method uses a combination of deep learning methods and a metaheuristic to provide an efficient model for the diagnosis of the foot fractures problem. the method has been first based on applying some preprocessing steps before using the model for the features extraction and classification of the problem. the main model is based on a pre-trained ZFNet. The final layers of the network have been substituted using an extreme learning machine (ELM) in its entirety. The ELM part also optimized based on a new developed metaheuristic, called enhanced snow ablation optimizer (ESAO), to achieve better results. for validating the effectiveness of the proposed ZFNet/ELM/ESAO-based model, it has been applied to a standard benchmark from Institutional Review Board (IRB) and the findings have been compared to some different high-tech methods, including Decision Tree / K-Nearest Neighbour (DT/KNN), Linear discriminant analysis (LDA), Inception-ResNet Faster R-CNN architecture (FRCNN), Transfer learning‑based ensemble convolutional neural network (TL-ECNN), and combined model containing a convolutional neural network and long short-term memory (DCNN/LSTM). Final results show that using the proposed ZFNet/ELM/ESAO-based can be utilized as an efficient model for the diagnosis of the foot fractures.

PMID:39558102 | DOI:10.1038/s41598-024-80132-8

Nat Methods. 2024 Nov 18. doi: 10.1038/s41592-024-02505-1. Online ahead of print.

ABSTRACT

While advances in single-particle cryo-EM have enabled the structural determination of macromolecular complexes at atomic resolution, particle orientation bias (the 'preferred' orientation problem) remains a complication for most specimens. Existing solutions have relied on biochemical and physical strategies applied to the specimen and are often complex and challenging. Here, we develop spIsoNet, an end-to-end self-supervised deep learning-based software to address map anisotropy and particle misalignment caused by the preferred-orientation problem. Using preferred-orientation views to recover molecular information in under-sampled views, spIsoNet improves both angular isotropy and particle alignment accuracy during 3D reconstruction. We demonstrate spIsoNet's ability to generate near-isotropic reconstructions from representative biological systems with limited views, including ribosomes, β-galactosidases and a previously intractable hemagglutinin trimer dataset. spIsoNet can also be generalized to improve map isotropy and particle alignment of preferentially oriented molecules in subtomogram averaging. Therefore, without additional specimen-preparation procedures, spIsoNet provides a general computational solution to the preferred-orientation problem.

PMID:39558095 | DOI:10.1038/s41592-024-02505-1

Sci Rep. 2024 Nov 18;14(1):28435. doi: 10.1038/s41598-024-79602-w.

ABSTRACT

Synthetic data is becoming a valuable tool for computational pathologists, aiding in tasks like data augmentation and addressing data scarcity and privacy. However, its use necessitates careful planning and evaluation to prevent the creation of clinically irrelevant artifacts.This manuscript introduces a comprehensive pipeline for generating and evaluating synthetic pathology data using a diffusion model. The pipeline features a multifaceted evaluation strategy with an integrated explainability procedure, addressing two key aspects of synthetic data use in the medical domain.The evaluation of the generated data employs an ensemble-like approach. The first step includes assessing the similarity between real and synthetic data using established metrics. The second step involves evaluating the usability of the generated images in deep learning models accompanied with explainable AI methods. The final step entails verifying their histopathological realism through questionnaires answered by professional pathologists. We show that each of these evaluation steps are necessary as they provide complementary information on the generated data's quality.The pipeline is demonstrated on the public GTEx dataset of 650 Whole Slide Images (WSIs), including five different tissues. An equal number of tiles from each tissue are generated and their reliability is assessed using the proposed evaluation pipeline, yielding promising results.In summary, the proposed workflow offers a comprehensive solution for generative AI in digital pathology, potentially aiding the community in their transition towards digitalization and data-driven modeling.

PMID:39557989 | DOI:10.1038/s41598-024-79602-w

Sci Rep. 2024 Nov 18;14(1):28496. doi: 10.1038/s41598-024-79271-9.

ABSTRACT

Assessing localization of the transient receptor potential vanilloid-1 (TRPV1) in skin nerve fibers is crucial for understanding its role in peripheral neuropathy and pain. However, information on the specificity and sensitivity of TRPV1 antibodies used for immunofluorescence (IF) on human skin is currently lacking. To find a reliable TRPV1 antibody and IF protocol, we explored antibody candidates from different manufacturers, used rat DRG sections and human skin samples for screening and human TRPV1-expressing HEK293 cells for further validation. Final specificity assessment was done on human skin samples. Additionally, we developed two automated image analysis methods: a Python-based deep-learning approach and a Fiji-based machine-learning approach. These methods involve training a model or classifier for nerve fibers based on pre-annotations and utilize a nerve fiber mask to filter and count TRPV1 immunoreactive puncta and TRPV1 fluorescence intensity on nerve fibers. Both automated analysis methods effectively distinguished TRPV1 signals on nerve fibers from those in keratinocytes, demonstrating high reliability as evidenced by excellent intraclass correlation coefficient (ICC) values exceeding 0.75. This method holds the potential to uncover alterations in TRPV1 associated with neuropathic pain conditions, using a minimally invasive approach.

PMID:39557902 | DOI:10.1038/s41598-024-79271-9

Sci Rep. 2024 Nov 18;14(1):28501. doi: 10.1038/s41598-024-77549-6.

ABSTRACT

Autism spectrum disorder is a developmental condition that affects the social and behavioral abilities of growing children. Early detection of autism spectrum disorder can help children to improve their cognitive abilities and quality of life. The research in the area of autism spectrum disorder reports that it can be detected from cognitive tests and physical activities of children. The present research reports on the detection of autism spectrum disorder from the facial attributes of children. Children with autism spectrum disorder show ambiguous facial expressions which are different from the facial attributes of normal children. To detect autism spectrum disorder from facial images, this work presents an improvised variant of the YOLOv7-tiny model. The presented model is developed by integrating a pyramid of dilated convolutional layers in the feature extraction network of the YOLOv7-tiny model. Further, its recognition abilities are enhanced by incorporating an additional YOLO detection head. The developed model can detect faces with the presence of autism features by drawing bounding boxes and confidence scores. The entire work has been carried out on a self-annotated autism face dataset. The developed model achieved a mAP value of 79.56% which was better than the baseline YOLOv7-tiny and state-of-the-art YOLOv8 Small model.

PMID:39557896 | DOI:10.1038/s41598-024-77549-6

Sci Data. 2024 Nov 18;11(1):1242. doi: 10.1038/s41597-024-04062-w.

ABSTRACT

We present detailed annual land cover maps for the Baltic Sea region, spanning more than two decades (2000-2022). The maps provide information on eighteen land cover (LC) classes, including eight general LC types, eight major crop types and grassland, and two peat bog-related classes. Our maps represent the first homogenized annual dataset for the region and address gaps in current land use and land cover products, such as a lack of detail on crop sequences and peat bog exploitation. To create the maps, we used annual multi-temporal remote sensing data combined with a data encoding structure and deep learning classification. We obtained the training data from publicly available open datasets. The maps were validated using independent field survey data from the Land Use/Cover Area Frame Survey (LUCAS) and expert annotations from high-resolution imagery. The quantitative and qualitative results of the maps provide a reliable data source for monitoring agricultural transformations, peat bog exploitation, and restoration activities in the Baltic Sea region and its surrounding countries.

PMID:39557873 | DOI:10.1038/s41597-024-04062-w

Comput Biol Med. 2024 Nov 17;184:109411. doi: 10.1016/j.compbiomed.2024.109411. Online ahead of print.

ABSTRACT

Brain age, an emerging biomarker for brain diseases and aging, is typically predicted using single-modality T1-weighted structural MRI data. This study investigates the benefits of integrating structural MRI with diffusion MRI to enhance brain age prediction. We propose an attention-based deep learning model that fuses global-context information from structural MRI with local details from diffusion metrics. The model was evaluated using two large datasets: the Human Connectome Project (HCP, n = 1064, age 22-37) and the Cambridge Center for Aging and Neuroscience (Cam-CAN, n = 639, age 18-88). It was tested for generalizability and robustness on three independent datasets (n = 546, age 20-86), reproducibility on a test-retest dataset (n = 44, age 22-35), and longitudinal consistency (n = 129, age 46-92). We also examined the relationship between predicted brain age and behavioral measures. Results showed that the multimodal model improved prediction accuracy, achieving mean absolute errors (MAEs) of 2.44 years in the HCP dataset (sagittal plane) and 4.36 years in the Cam-CAN dataset (axial plane). The corresponding R2 values were 0.258 and 0.914, respectively, reflecting the model's ability to explain variance in the predictions across both datasets. Compared to single-modality models, the multimodal approach showed better generalization, reducing MAEs by 10-76 % and enhancing robustness by 22-82 %. While the multimodal model exhibited superior reproducibility, the sMRI model showed slightly better longitudinal consistency. Importantly, the multimodal model revealed unique associations between predicted brain age and behavioral measures, such as walking endurance and loneliness in the HCP dataset, which were not detected with chronological age alone. In the Cam-CAN dataset, brain age and chronological age exhibited similar correlations with behavioral measures. By integrating sMRI and dMRI through an attention-based model, our proposed approach enhances predictive accuracy and provides deeper insights into the relationship between brain aging and behavior.

PMID:39556917 | DOI:10.1016/j.compbiomed.2024.109411

Proc Natl Acad Sci U S A. 2024 Nov 26;121(48):e2402028121. doi: 10.1073/pnas.2402028121. Epub 2024 Nov 18.

ABSTRACT

The loss of phosphorous (P) from the land to aquatic systems has polluted waters and threatened food production worldwide. Systematic trend analysis of P, a nonrenewable resource, has been challenging, primarily due to sparse and inconsistent historical data. Here, we leveraged intensive hydrometeorological data and the recent renaissance of deep learning approaches to fill data gaps and reconstruct temporal trends. We trained a multitask long short-term memory model for total P (TP) using data from 430 rivers across the contiguous United States (CONUS). Trend analysis of reconstructed daily records (1980-2019) shows widespread decline in concentrations, with declining, increasing, and insignificantly changing trends in 60%, 28%, and 12% of the rivers, respectively. Concentrations in urban rivers have declined the most despite rising urban population in the past decades; concentrations in agricultural rivers however have mostly increased, suggesting not-as-effective controls of nonpoint sources in agriculture lands compared to point sources in cities. TP loss, calculated as fluxes by multiplying concentration and discharge, however exhibited an overall increasing rate of 6.5% per decade at the CONUS scale over the past 40 y, largely due to increasing river discharge. Results highlight the challenge of reducing TP loss that is complicated by changing river discharge in a warming climate.

PMID:39556745 | DOI:10.1073/pnas.2402028121

PLoS One. 2024 Nov 18;19(11):e0312865. doi: 10.1371/journal.pone.0312865. eCollection 2024.

ABSTRACT

To understand and measure political information consumption in the high-choice media environment, we need new methods to trace individual interactions with online content and novel techniques to analyse and detect politics-related information. In this paper, we report the results of a comparative analysis of the performance of automated content analysis techniques for detecting political content in the German language across different platforms. Using three validation datasets, we compare the performance of three groups of detection techniques relying on dictionaries, classic supervised machine learning, and deep learning. We also examine the impact of different modes of data preprocessing on the low-cost implementations of these techniques using a large set (n = 66) of models. Our results show the limited impact of preprocessing on model performance, with the best results for less noisy data being achieved by deep learning- and classic machine learning-based models, in contrast to the more robust performance of dictionary-based models on noisy data.

PMID:39556542 | DOI:10.1371/journal.pone.0312865

Neural Comput. 2024 Nov 18:1-34. doi: 10.1162/neco_a_01726. Online ahead of print.

ABSTRACT

In recent years, models based on the transformer architecture have seen widespread applications and have become one of the core tools in the field of deep learning. Numerous successful and efficient techniques, such as parameter-efficient fine-tuning and efficient scaling, have been proposed surrounding their applications to further enhance performance. However, the success of these strategies has always lacked the support of rigorous mathematical theory. To study the underlying mechanisms behind transformers and related techniques, we first propose a transformer learning framework motivated by distribution regression, with distributions being inputs, connect a two-stage sampling process with natural language processing, and present a mathematical formulation of the attention mechanism called attention operator. We demonstrate that by the attention operator, transformers can compress distributions into function representations without loss of information. Moreover, with the advantages of our novel attention operator, transformers exhibit a stronger capability to learn functionals with more complex structures than convolutional neural networks and fully connected networks. Finally, we obtain a generalization bound within the distribution regression framework. Throughout theoretical results, we further discuss some successful techniques emerging with large language models (LLMs), such as prompt tuning, parameter-efficient fine-tuning, and efficient scaling. We also provide theoretical insights behind these techniques within our novel analysis framework.

PMID:39556504 | DOI:10.1162/neco_a_01726

Skeletal Radiol. 2024 Nov 18. doi: 10.1007/s00256-024-04818-w. Online ahead of print.

ABSTRACT

OBJECTIVE: To determine an optimal region of interest (ROI) for opportunistic screening of osteoporosis in terms of short-term in vivo diagnostic precision.

MATERIALS AND METHODS: We included patients who underwent two CT scans and one dual-energy X-ray absorptiometry scan within a month in 2022. Deep-learning software automatically measured the attenuation in L1 using 54 ROIs (three slice thicknesses × six shapes × three intravertebral levels). To identify factors associated with a lower attenuation difference between the two CT scans, mixed-effect model analysis was performed with ROI-level (slice thickness, shape, intravertebral levels) and patient-level (age, sex, patient diameter, change in CT machine) factors. The root-mean-square standard deviation (RMSSD) and area under the receiver-operating-characteristic curve (AUROC) were calculated.

RESULTS: In total, 73 consecutive patients (mean age ± standard deviation, 69 ± 9 years, 38 women) were included. A lower attenuation difference was observed in ROIs in images with slice thicknesses of 1 and 3 mm than that in images with a slice thickness of 5 mm (p < .001), in large elliptical ROIs (p = .007 or < .001, respectively), and in mid- or cranial-level ROIs than that in caudal-level ROIs (p < .001). No patient-level factors were significantly associated with the attenuation difference. Large, elliptical ROIs placed at the mid-level of L1 on images with 1- or 3-mm slice thicknesses yielded RMSSDs of 12.4-12.5 HU and AUROCs of 0.90.

CONCLUSION: The largest possible regions of interest drawn in the mid-level trabecular portion of the L1 vertebra on thin-slice images may yield improvements in the precision of opportunistic screening for osteoporosis via CT.

PMID:39556270 | DOI:10.1007/s00256-024-04818-w

Endocrine. 2024 Nov 18. doi: 10.1007/s12020-024-04091-w. Online ahead of print.

ABSTRACT

BACKGROUND: Cervical lymph node metastasis (CLNM) is the most common form of thyroid cancer metastasis. Accurate preoperative CLNM diagnosis is of more importance in patients with papillary thyroid cancer (PTC). However, there is currently no unified methods to objectively predict CLNM risk from ultrasonography in PTC patients.This study aimed to develop a deep learning (DL) model to help clinicians more accurately determine the existence of CLNM risk in patients with PTC and then assist them with treatment decisions.

METHODS: Ultrasound dynamic videos of 388 patients with 717 thyroid nodules were retrospectively collected from Zhejiang Cancer Hospital between January 2020 and June 2022. Five deep learning (DL) models were investigated to examine its efficacy for predicting CLNM risks and their performances were also compared with those predicted using two-dimensional ultrasound static images.

RESULTS: In the testing dataset (n = 78), the DenseNet121 model trained on ultrasound dynamic videos outperformed the other four DL models as well as the DL model trained using the two-dimensional (2D) static images across all metrics. Specifically, using DenseNet121, the comparison between the 3D model and 2D model for all metrics are shown as below: AUROC: 0.903 versus 0.828, sensitivity: 0.877 versus 0.871, specificity: 0.865 versus 0.659.

CONCLUSIONS: This study demonstrated that the DenseNet121 model has the greatest potential in distinguishing CLNM from non-CLNM in patients with PTC. Dynamic videos also offered more information about the disease states which have proven to be more efficient and robust in identifying CLNM compared to statis images.

PMID:39556263 | DOI:10.1007/s12020-024-04091-w

Pediatr Radiol. 2024 Nov 18. doi: 10.1007/s00247-024-06098-x. Online ahead of print.

ABSTRACT

Artificial intelligence (AI) is increasingly recognized for its transformative potential in radiology; yet, its application in pediatric radiology is relatively limited when compared to the whole of radiology. This manuscript introduces pediatric radiologists to essential AI concepts, including topics such as use case, data science, machine learning, deep learning, natural language processing, and generative AI as well as basics of AI training and validating. We outline the unique challenges of applying AI in pediatric imaging, such as data scarcity and distinct clinical characteristics, and discuss the current uses of AI in pediatric radiology, including both image interpretive and non-interpretive tasks. With this overview, we aim to equip pediatric radiologists with the foundational knowledge needed to engage with AI tools and inspire further exploration and innovation in the field.

PMID:39556194 | DOI:10.1007/s00247-024-06098-x

Eur Radiol Exp. 2024 Nov 18;8(1):129. doi: 10.1186/s41747-024-00527-0.

ABSTRACT

BACKGROUND: The presence of a blurred area, depending on its localization, in a mammogram can limit diagnostic accuracy. The goal of this study was to develop a model for automatic detection of blur in diagnostically relevant locations in digital mammography.

METHODS: A retrospective dataset consisting of 152 examinations acquired with mammography machines from three different vendors was utilized. The blurred areas were contoured by expert breast radiologists. Normalized Wiener spectra (nWS) were extracted in a sliding window manner from each mammogram. These spectra served as input for a convolutional neural network (CNN) generating the probability of the spectra originating from a blurred region. The resulting blur probability mask, upon thresholding, facilitated the classification of a mammogram as either blurred or sharp. Ground truth for the test set was defined by the consensus of two radiologists.

RESULTS: A significant correlation between the view (p < 0.001), as well as between the laterality and the presence of blur (p = 0.004) was identified. The developed model AUROC of 0.808 (95% confidence interval 0.794-0.821) aligned with the consensus in 78% (67-83%) of mammograms classified as blurred. For mammograms classified by consensus as sharp, the model achieved agreement in 75% (67-83%) of them.

CONCLUSION: A model for blur detection was developed and assessed. The results indicate that a robust approach to blur detection, based on feature extraction in frequency space, tailored to radiologist expertise regarding clinical relevance, could eliminate the subjectivity associated with the visual assessment.

RELEVANCE STATEMENT: This blur detection model, if implemented in clinical practice, could provide instantaneous feedback to technicians, allowing for prompt mammogram retakes and ensuring that only high-quality mammograms are sent for screening and diagnostic tasks.

KEY POINTS: Blurring in mammography limits radiologist interpretation and diagnostic accuracy. This objective blur detection tool ensures image quality, and reduces retakes and unnecessary exposures. Wiener spectrum analysis and CNN enabled automated blur detection in mammography.

PMID:39556167 | DOI:10.1186/s41747-024-00527-0

Int J Legal Med. 2024 Nov 18. doi: 10.1007/s00414-024-03370-5. Online ahead of print.

ABSTRACT

This article evaluates the criteria for diatom testing in forensic investigations, focusing on drowning cases. Diatoms, unicellular algae found in aquatic environments, are critical to the determination of drowning because water containing diatoms is inhaled during submersion. The primary objectives include defining the exact amount and type of tissue to be analyzed, expressed in terms of diatom concentration relative to tissue weight, and detailing the conditions under which water samples are collected to study the diatom flora at the site. In addition, the importance of accurately identifying diatom taxa and comparing them by unit weight is emphasized. To improve the reliability of diatom testing, the study discusses advanced methods such as microwave digestion, vacuum filtration, and automated scanning electron microscopy (MD-VF-Auto SEM), which offer higher sensitivity and specificity. The integration of DNA sequencing and deep learning techniques is explored, offering promising improvements in diatom detection and classification. These advances aim to reduce false positives and improve the accuracy of determining drowning as the cause of death. The article highlights the need for standardized protocols for diatom testing to ensure consistency and reliability. By incorporating new technologies and refining existing methods, the forensic application of diatom testing can be significantly improved, allowing for more accurate and reliable conclusions in drowning investigations.

PMID:39556128 | DOI:10.1007/s00414-024-03370-5

J Hand Surg Am. 2024 Nov 15:S0363-5023(24)00432-5. doi: 10.1016/j.jhsa.2024.09.008. Online ahead of print.

ABSTRACT

PURPOSE: Fragility fractures associated with osteoporosis and osteopenia are a common cause of morbidity and mortality. Current methods of diagnosing low bone mineral density require specialized dual x-ray absorptiometry (DXA) scans. Plain hand radiographs may have utility as an alternative screening tool, although optimal diagnostic radiographic parameters are unknown, and measurement is prone to human error. The aim of the present study was to develop and validate an artificial intelligence algorithm to screen for osteoporosis and osteopenia using standard hand radiographs.

METHODS: A cohort of patients with both a DXA scan and a plain hand radiograph within 12 months of one another was identified. Hand radiographs were labeled as normal, osteopenia, or osteoporosis based on corresponding DXA hip T-scores. A deep learning algorithm was developed using the ResNet-50 framework and trained to predict the presence of osteoporosis or osteopenia on hand radiographs using labeled images. The results from the algorithm were validated using a separate balanced validation set, with the calculation of sensitivity, specificity, accuracy, and receiver operating characteristic curve using definitions from corresponding DXA scans as the reference standard.

RESULTS: There was a total of 687 images in the normal category, 607 images in the osteopenia category, and 130 images in the osteoporosis category for a total of 1,424 images. When predicting low bone density (osteopenia or osteoporosis) versus normal bone density, sensitivity was 88.5%, specificity was 65.4%, overall accuracy was 80.8%, and the area under the curve was 0.891, at the standard threshold of 0.5. If optimizing for both sensitivity and specificity, at a threshold of 0.655, the model achieved a sensitivity of 84.6% at a specificity of 84.6%.

CONCLUSIONS: The findings represent a possible step toward more accessible, cost-effective, automated diagnosis and therefore earlier treatment of osteoporosis/osteopenia.

TYPE OF STUDY/LEVEL OF EVIDENCE: Diagnostic II.

PMID:39556066 | DOI:10.1016/j.jhsa.2024.09.008