Med Phys. 2024 Sep 24. doi: 10.1002/mp.17416. Online ahead of print.

ABSTRACT

BACKGROUND: Treatment planning is currently a patient specific, time-consuming, and resource demanding task in radiotherapy. Dose-volume histogram (DVH) prediction plays a critical role in automating this process. The geometric relationship between DVHs in radiotherapy plans and organs-at-risk (OAR) and planning target volume (PTV) has been well established. This study explores the potential of deep learning models for predicting DVHs using images and subsequent human intervention facilitated by a large-language model (LLM) to enhance the planning quality.

METHOD: We propose a pipeline to convert unstructured images to a structured graph consisting of image-patch nodes and dose nodes. A novel Dose Graph Neural Network (DoseGNN) model is developed for predicting DVHs from the structured graph. The proposed DoseGNN is enhanced with the LLM to encode massive knowledge from prescriptions and interactive instructions from clinicians. In this study, we introduced an online human-AI collaboration (OHAC) system as a practical implementation of the concept proposed for the automation of intensity-modulated radiotherapy (IMRT) planning.

RESULTS: The proposed DoseGNN model was compared to widely employed DL models used in radiotherapy, including Swin Transformer, 3D U-Net CNN, and vanilla MLP. For PTV, DoseGNN achieved the mean absolute error (MAE) of D m a x ${D}_{max}$ , D m e a n ${D}_{mean}$ , D 95 ${D}_{95}$ , and D 1 ${D}_1$ between true plans and predicted plans that were 64%, 53%, 64%, 61% of the best baseline model. For the worst case among OARs (left lung, right lung, chest wall, heart, spinal cord), DoseGNN achieved the mean absolute error of D m a x ${D}_{max}$ , D m e a n ${D}_{mean}$ , D 50 ${D}_{50}$ that were 85%, 91%, 80% of the best baseline model. Moreover, the LLM-empowered DoseGNN model facilitates seamless adjustment to treatment plans through interaction with clinicians using natural language.

CONCLUSION: We developed DoseGNN, a novel deep learning model for predicting delivered radiation doses from medical images, enhanced by LLM to allow adjustment through seamless interaction with clinicians. The preliminary results confirm DoseGNN's superior accuracy in DVH prediction relative to typical DL methods, highlighting its potential to facilitate an online clinician-AI collaboration system for streamlined treatment planning automation.

PMID:39316523 | DOI:10.1002/mp.17416

IEEE/ACM Trans Comput Biol Bioinform. 2024 Sep 24;PP. doi: 10.1109/TCBB.2024.3467261. Online ahead of print.

ABSTRACT

Accurate identification of antifreeze proteins (AFPs) is crucial in developing biomimetic synthetic anti-icing materials and low-temperature organ preservation materials. Although numerous machine learning-based methods have been proposed for AFPs prediction, the complex and diverse nature of AFPs limits the prediction performance of existing methods. In this study, we propose AFP-Deep, a new deep learning method to predict antifreeze proteins by integrating embedding from protein sequences with pre-trained protein language models and evolutionary contexts with hybrid feature extraction networks. The experimental results demonstrated that the main advantage of AFP-Deep is its utilization of pre-trained protein language models, which can extract discriminative global contextual features from protein sequences. Additionally, the hybrid deep neural networks designed for protein language models and evolutionary context feature extraction enhance the correlation between embeddings and antifreeze pattern. The performance evaluation results show that AFP-Deep achieves superior performance compared to state-of-the-art models on benchmark datasets, achieving an AUPRC of 0.724 and 0.924, respectively.

PMID:39316498 | DOI:10.1109/TCBB.2024.3467261

IEEE Trans Neural Netw Learn Syst. 2024 Sep 24;PP. doi: 10.1109/TNNLS.2024.3453973. Online ahead of print.

ABSTRACT

Despite the significant advances achieved in deep learning, the deep neural networks' (DNNs) design approach remains notoriously tedious, depending primarily on intuition, experience, and trial and error. This human-dependent process is often time-consuming and prone to errors. Furthermore, the models are generally bound to their training contexts, with no considerations to their surrounding environments. Continual adaptiveness and automation of neural networks is of paramount importance to several domains where model accessibility is limited after deployment (e.g., IoT devices, self-driving vehicles, etc.). Additionally, even accessible models require frequent maintenance postdeployment to overcome issues such as data/concept drift, which can be cumbersome and restrictive. By leveraging and combining approaches from neural architecture search (NAS) and continual learning (CL), more robust and adaptive agents can be developed. This study conducts the first extensive review on the intersection between NAS and CL, formalizing the prospective paradigm and outlining research directions for lifelong autonomous DNNs.

PMID:39316489 | DOI:10.1109/TNNLS.2024.3453973

IEEE J Biomed Health Inform. 2024 Sep 24;PP. doi: 10.1109/JBHI.2024.3467090. Online ahead of print.

ABSTRACT

Motor imagery (MI) decoding is the basis of external device control via electroencephalogram (EEG). However, the majority of studies prioritize enhancing the accuracy of decoding methods, often overlooking the magnitude and computational resource demands of deep learning models. In this study, we propose a novel lightweight Multi-Scale Feature Residual Convolutional Neural Network (MFRC-Net). MFRC-Net primarily consists of two blocks: temporal multi-scale residual convolution blocks and cross-domain dual-stream spatial convolution blocks. The former captures dynamic changes in EEG signals across various time scales through multi-scale grouped convolution and backbone temporal convolution skip connections; the latter improves local spatial feature extraction and calibrates feature mapping through the introduction of cross-domain spatial filtering layers. Furthermore, by specifically optimizing the loss function, MFRC-Net effectively reduces sensitivity to outliers. Experiment results on the BCI Competition IV 2a dataset and the SHU dataset demonstrate that, with a parameter size of only 13K, MFRC-Net achieves accuracy of 85.1% and 69.3%, respectively, surpassing current state-of-the-art models. The integration of temporal multi-scale residual convolution blocks and crossdomain dual-stream spatial convolution blocks in lightweight models significantly boosts performance, as evidenced by ablation studies and visualizations.

PMID:39316474 | DOI:10.1109/JBHI.2024.3467090

JMIR Aging. 2024 Sep 24;7:e57926. doi: 10.2196/57926.

ABSTRACT

BACKGROUND: The severity of Alzheimer disease and related dementias (ADRD) is rarely documented in structured data fields in electronic health records (EHRs). Although this information is important for clinical monitoring and decision-making, it is often undocumented or "hidden" in unstructured text fields and not readily available for clinicians to act upon.

OBJECTIVE: We aimed to assess the feasibility and potential bias in using keywords and rule-based matching for obtaining information about the severity of ADRD from EHR data.

METHODS: We used EHR data from a large academic health care system that included patients with a primary discharge diagnosis of ADRD based on ICD-9 (International Classification of Diseases, Ninth Revision) and ICD-10 (International Statistical Classification of Diseases, Tenth Revision) codes between 2014 and 2019. We first assessed the presence of ADRD severity information and then the severity of ADRD in the EHR. Clinicians' notes were used to determine the severity of ADRD based on two criteria: (1) scores from the Mini Mental State Examination and Montreal Cognitive Assessment and (2) explicit terms for ADRD severity (eg, "mild dementia" and "advanced Alzheimer disease"). We compiled a list of common ADRD symptoms, cognitive test names, and disease severity terms, refining it iteratively based on previous literature and clinical expertise. Subsequently, we used rule-based matching in Python using standard open-source data analysis libraries to identify the context in which specific words or phrases were mentioned. We estimated the prevalence of documented ADRD severity and assessed the performance of our rule-based algorithm.

RESULTS: We included 9115 eligible patients with over 65,000 notes from the providers. Overall, 22.93% (2090/9115) of patients were documented with mild ADRD, 20.87% (1902/9115) were documented with moderate or severe ADRD, and 56.20% (5123/9115) did not have any documentation of the severity of their ADRD. For the task of determining the presence of any ADRD severity information, our algorithm achieved an accuracy of >95%, specificity of >95%, sensitivity of >90%, and an F1-score of >83%. For the specific task of identifying the actual severity of ADRD, the algorithm performed well with an accuracy of >91%, specificity of >80%, sensitivity of >88%, and F1-score of >92%. Comparing patients with mild ADRD to those with more advanced ADRD, the latter group tended to contain older, more likely female, and Black patients, and having received their diagnoses in primary care or in-hospital settings. Relative to patients with undocumented ADRD severity, those with documented ADRD severity had a similar distribution in terms of sex, race, and rural or urban residence.

CONCLUSIONS: Our study demonstrates the feasibility of using a rule-based matching algorithm to identify ADRD severity from unstructured EHR report data. However, it is essential to acknowledge potential biases arising from differences in documentation practices across various health care systems.

PMID:39316421 | DOI:10.2196/57926

Jpn J Radiol. 2024 Sep 24. doi: 10.1007/s11604-024-01666-5. Online ahead of print.

ABSTRACT

PURPOSE: To evaluate deep learning-reconstructed (DLR)-fluid-attenuated inversion recovery (FLAIR) images generated from undersampled data, compare them with fully sampled and rapidly acquired FLAIR images, and assess their potential for white matter hyperintensity evaluation.

MATERIALS AND METHODS: We examined 30 patients with white matter hyperintensities, obtaining fully sampled FLAIR images (standard FLAIR, std-FLAIR). We created accelerated FLAIR (acc-FLAIR) images using one-third of the fully sampled data and applied deep learning to generate DLR-FLAIR images. Three neuroradiologists assessed the quality (amount of noise and gray/white matter contrast) in all three image types. The reproducibility of hyperintensities was evaluated by comparing a subset of 100 hyperintensities in acc-FLAIR and DLR-FLAIR images with those in the std-FLAIR images. Quantitatively, similarities and errors of the entire image and the focused regions on white matter hyperintensities in acc-FLAIR and DLR-FLAIR images were measured against std-FLAIR images using structural similarity index measure (SSIM), regional SSIM, normalized root mean square error (NRMSE), and regional NRMSE values.

RESULTS: All three neuroradiologists evaluated DLR-FLAIR as having significantly less noise and higher image quality scores compared with std-FLAIR and acc-FLAIR (p < 0.001). All three neuroradiologists assigned significantly higher frontal lobe gray/white matter visibility scores for DLR-FLAIR than for acc-FLAIR (p < 0.001); two neuroradiologists attributed significantly higher scores for DLR-FLAIR than for std-FLAIR (p < 0.05). Regarding white matter hyperintensities, all three neuroradiologists significantly preferred DLR-FLAIR (p < 0.0001). DLR-FLAIR exhibited higher similarity to std-FLAIR in terms of visibility of the hyperintensities, with 97% of the hyperintensities rated as nearly identical or equivalent. Quantitatively, DLR-FLAIR demonstrated significantly higher SSIM and regional SSIM values than acc-FLAIR, with significantly lower NRMSE and regional NRMSE values (p < 0.0001).

CONCLUSIONS: DLR-FLAIR can reduce scan time and generate images of similar quality to std-FLAIR in patients with white matter hyperintensities. Therefore, DLR-FLAIR may serve as an effective method in traditional magnetic resonance imaging protocols.

PMID:39316286 | DOI:10.1007/s11604-024-01666-5

Plant Mol Biol. 2024 Sep 24;114(5):106. doi: 10.1007/s11103-024-01500-6.

ABSTRACT

Photosynthetic proteins play a crucial role in agricultural productivity by harnessing light energy for plant growth. Understanding these proteins, especially within C3 and C4 pathways, holds promise for improving crops in challenging environments. Despite existing models, a comprehensive computational framework specifically targeting plant photosynthetic proteins is lacking. The underutilization of plant datasets in computational algorithms accentuates the gap this study aims to fill by introducing a novel sequence-based computational method for identifying these proteins. The scope of this study encompassed diverse plant species, ensuring comprehensive representation across C3 and C4 pathways. Utilizing six deep learning models and seven shallow learning algorithms, paired with six sequence-derived feature sets followed by feature selection strategy, this study developed a comprehensive model for prediction of plant-specific photosynthetic proteins. Following 5-fold cross-validation analysis, LightGBM with 65 and 90 LGBM-VIM selected features respectively emerged as the best models for C3 (auROC: 91.78%, auPRC: 92.55%) and C4 (auROC: 99.05%, auPRC: 99.18%) plants. Validation using an independent dataset confirmed the robustness of the proposed model for both C3 (auROC: 87.23%, auPRC: 88.40%) and C4 (auROC: 92.83%, auPRC: 92.29%) categories. Comparison with existing methods demonstrated the superiority of the proposed model in predicting plant-specific photosynthetic proteins. This study further established a free online prediction server PredPSP ( https://iasri-sg.icar.gov.in/predpsp/ ) to facilitate ongoing efforts for identifying photosynthetic proteins in C3 and C4 plants. Being first of its kind, this study offers valuable insights into predicting plant-specific photosynthetic proteins which holds significant implications for plant biology.

PMID:39316155 | DOI:10.1007/s11103-024-01500-6

Anal Bioanal Chem. 2024 Sep 24. doi: 10.1007/s00216-024-05550-z. Online ahead of print.

ABSTRACT

Exosome analysis plays pivotal roles in various physiological and pathological processes. Plasmonic scattering microscopy (PSM) has proven to be an excellent label-free imaging platform for exosome detection. However, accurately detecting images scattered from exosomes remains a challenging task due to noise interference. Herein, we proposed an image processing strategy based on a new blind super-resolution deep learning neural network, named ESRGAN-SE, to improve the resolution of exosome PSI images. This model can obtain super-resolution reconstructed images without increasing experimental complexity. The trained model can directly generate high-resolution plasma scattering images from low-resolution images collected in experiments. The results of experiments involving the detection of light scattered by exosomes showed that the proposed super-resolution detection method has strong generalizability and robustness. Moreover, ESRGAN-SE achieved excellent results of 35.52036, 0.09081, and 8.13176 in terms of three reference-free image quality assessment metrics, respectively. These results show that the proposed network can effectively reduce image information loss, enhance mutual information between pixels, and decrease feature differentiation. And, the single-image SNR evaluation score of 3.93078 also showed that the distinction between the target and the background was significant. The suggested model lays the foundation for a potentially successful approach to imaging analysis. This approach has the potential to greatly improve the accuracy and efficiency of exosome analysis, leading to more accurate cancer diagnosis and potentially improving patient outcomes.

PMID:39316091 | DOI:10.1007/s00216-024-05550-z

Neuroradiology. 2024 Sep 24. doi: 10.1007/s00234-024-03461-5. Online ahead of print.

ABSTRACT

PURPOSE: The time-intensive nature of acquiring 3D T1-weighted MRI and analyzing brain volumetry limits quantitative evaluation of brain atrophy. We explore the feasibility and reliability of deep learning-based accelerated MRI scans for brain volumetry.

METHODS: This retrospective study collected 3D T1-weighted data using 3T from 42 participants for the simulated acceleration dataset and 48 for the validation dataset. The simulated acceleration dataset consists of three sets at different simulated acceleration levels (Simul-Accel) corresponding to level 1 (65% undersampling), 2 (70%), and 3 (75%). These images were then subjected to deep learning-based reconstruction (Simul-Accel-DL). Conventional images (Conv) without acceleration and DL were set as the reference. In the validation dataset, DICOM images were collected from Conv and accelerated scan with DL-based reconstruction (Accel-DL). The image quality of Simul-Accel-DL was evaluated using quantitative error metrics. Volumetric measurements were evaluated using intraclass correlation coefficients (ICCs) and linear regression analysis in both datasets. The volumes were estimated by two software, NeuroQuant and DeepBrain.

RESULTS: Simul-Accel-DL across all acceleration levels revealed comparable or better error metrics than Simul-Accel. In the simulated acceleration dataset, ICCs between Conv and Simul-Accel-DL in all ROIs exceeded 0.90 for volumes and 0.77 for normative percentiles at all acceleration levels. In the validation dataset, ICCs for volumes > 0.96, ICCs for normative percentiles > 0.89, and R2 > 0.93 at all ROIs except pallidum demonstrated good agreement in both software.

CONCLUSION: DL-based reconstruction achieves clinical feasibility of 3D T1 brain volumetric MRI by up to 75% acceleration relative to full-sampled acquisition.

PMID:39316090 | DOI:10.1007/s00234-024-03461-5

Int Forum Allergy Rhinol. 2024 Sep 24. doi: 10.1002/alr.23458. Online ahead of print.

ABSTRACT

AI-enabled augmentation of nasal endoscopy video images is feasible in the clinical setting. Edge computing hardware can interface with existing nasal endoscopy equipment. Real-time AI performance can achieve an acceptable balance of accuracy and efficiency.

PMID:39316067 | DOI:10.1002/alr.23458

OMICS. 2024 Sep 24. doi: 10.1089/omi.2024.0154. Online ahead of print.

ABSTRACT

Identifying genomic regions under selection is essential for understanding the genetic mechanisms driving species evolution and adaptation. Traditional methods often fall short in detecting complex, spatially varying selection signals. Recent advances in deep learning, however, present promising new approaches for uncovering subtle selection signals that traditional methods might miss. In this study, we utilized the deep learning framework DeepGenomeScan to detect spatially varying selection signatures across 15 bovine populations worldwide. Our analysis uncovered novel insights into selective sweep hotspots within the bovine genome, revealing key genes associated with physiological and adaptive traits that were previously undetected. We identified significant quantitative trait loci linked to milk protein and fat percentages. By comparing the selection signatures identified in this study with those reported in the Bovine Genome Variation Database, we discovered 38 novel genes under selection that were not identified through traditional methods. These genes are primarily associated with milk and meat yield and quality. Our findings enhance our understanding of spatially varying selection's impact on bovine genomic diversity, laying a foundation for future research in genetic improvement and conservation. This is the first deep learning-based study of selection signatures in cattle, offering new insights for evolutionary and livestock genomics research.

PMID:39315920 | DOI:10.1089/omi.2024.0154

Development. 2024 Sep 15;151(18):dev204372. doi: 10.1242/dev.204372. Epub 2024 Sep 24.

ABSTRACT

During wound healing, epithelial cells divide, migrate and change shape. However, it is unclear how much these different cell behaviours contribute to the closure of the wound. A new paper in Development applies deep learning models to analyse large imaging datasets from the Drosophila pupa and quantify the contributions of the different cell behaviours. To learn more about the story behind the paper, we caught up with first author Jake Turley. Jake was a PhD student in Bristol, UK, but is now a Research Fellow at the Mechanobiology Institute, National University of Singapore.

PMID:39315487 | DOI:10.1242/dev.204372

Front Plant Sci. 2024 Sep 9;15:1365275. doi: 10.3389/fpls.2024.1365275. eCollection 2024.

ABSTRACT

Bird invasion will reduce the yield of high-value crops, which threatens the healthy development of agricultural economy. Sonic bird repellent has the advantages of large range, no time and geographical restrictions, and low cost, which has attracted people's attention in the field of agriculture. At present, there are few studies on the application of sonic bird repellents in pear orchards to minimize economic losses and prolong the adaptive capacity of birds. In this paper, a sound wave bird repellent system based on computer vision is designed, which combines deep learning target recognition technology to accurately identify birds and drive them away. The neural network model that can recognize birds is first trained and deployed to the server. Live video is captured by an installed webcam, and the sonic bird repellent is powered by an ESP-8266 relay switch. In a pear orchard, two experimental areas were divided into two experimental areas to test the designed sonic bird repellent device, and the number of bad fruits pecked by birds was used as an indicator to evaluate the bird repelling effect. The results showed that the pear pecked fruit rate was 6.03% in the pear orchard area that used the acoustic bird repeller based on computer recognition, 7.29% in the pear orchard area of the control group that used the acoustic bird repeller with continuous operation, and 13.07% in the pear orchard area that did not use any bird repellent device. While acoustic bird repellers based on computer vision can be more effective at repelling birds, they can be used in combination with methods such as fruit bags to reduce the economic damage caused by birds.

PMID:39315369 | PMC:PMC11416946 | DOI:10.3389/fpls.2024.1365275

Front Artif Intell. 2024 Sep 9;7:1376546. doi: 10.3389/frai.2024.1376546. eCollection 2024.

ABSTRACT

BACKGROUND: This study delves into the crucial domain of sperm segmentation, a pivotal component of male infertility diagnosis. It explores the efficacy of diverse architectural configurations coupled with various encoders, leveraging frames from the VISEM dataset for evaluation.

METHODS: The pursuit of automated sperm segmentation led to the examination of multiple deep learning architectures, each paired with distinct encoders. Extensive experimentation was conducted on the VISEM dataset to assess their performance.

RESULTS: Our study evaluated various deep learning architectures with different encoders for sperm segmentation using the VISEM dataset. While each model configuration exhibited distinct strengths and weaknesses, UNet++ with ResNet34 emerged as a top-performing model, demonstrating exceptional accuracy in distinguishing sperm cells from non-sperm cells. However, challenges persist in accurately identifying closely adjacent sperm cells. These findings provide valuable insights for improving automated sperm segmentation in male infertility diagnosis.

DISCUSSION: The study underscores the significance of selecting appropriate model combinations based on specific diagnostic requirements. It also highlights the challenges related to distinguishing closely adjacent sperm cells.

CONCLUSION: This research advances the field of automated sperm segmentation for male infertility diagnosis, showcasing the potential of deep learning techniques. Future work should aim to enhance accuracy in scenarios involving close proximity between sperm cells, ultimately improving clinical sperm analysis.

PMID:39315244 | PMC:PMC11418809 | DOI:10.3389/frai.2024.1376546

Med Biol Eng Comput. 2024 Sep 18. doi: 10.1007/s11517-024-03190-0. Online ahead of print.

ABSTRACT

Atherosclerosis causes heart disease by forming plaques in arterial walls. IVUS imaging provides a high-resolution cross-sectional view of coronary arteries and plaque morphology. Healthcare professionals diagnose and quantify atherosclerosis physically or using VH-IVUS software. Since manual or VH-IVUS software-based diagnosis is time-consuming, automated plaque characterization tools are essential for accurate atherosclerosis detection and classification. Recently, deep learning (DL) and computer vision (CV) approaches are promising tools for automatically classifying plaques on IVUS images. With this motivation, this manuscript proposes an automated atherosclerotic plaque classification method using a hybrid Ant Lion Optimizer with Deep Learning (AAPC-HALODL) technique on IVUS images. The AAPC-HALODL technique uses the faster regional convolutional neural network (Faster RCNN)-based segmentation approach to identify diseased regions in the IVUS images. Next, the ShuffleNet-v2 model generates a useful set of feature vectors from the segmented IVUS images, and its hyperparameters can be optimally selected by using the HALO technique. Finally, an average ensemble classification process comprising a stacked autoencoder (SAE) and deep extreme learning machine (DELM) model can be utilized. The MICCAI Challenge 2011 dataset was used for AAPC-HALODL simulation analysis. A detailed comparative study showed that the AAPC-HALODL approach outperformed other DL models with a maximum accuracy of 98.33%, precision of 97.87%, sensitivity of 98.33%, and F score of 98.10%.

PMID:39292382 | DOI:10.1007/s11517-024-03190-0

Med Biol Eng Comput. 2024 Sep 18. doi: 10.1007/s11517-024-03189-7. Online ahead of print.

ABSTRACT

Accurate and fast extraction of step parameters from video recordings of gait allows for richer information to be obtained from clinical tests such as Timed Up and Go. Current deep-learning methods are promising, but lack in accuracy for many clinical use cases. Extracting step parameters will often depend on extracted landmarks (keypoints) on the feet. We hypothesize that such keypoints can be determined with an accuracy relevant for clinical practice from video recordings by combining an existing general-purpose pose estimation method (OpenPose) with custom convolutional neural networks (convnets) specifically trained to identify keypoints on the heel. The combined method finds keypoints on the posterior and lateral aspects of the heel of the foot in side-view and frontal-view images from which step length and step width can be determined for calibrated cameras. Six different candidate convnets were evaluated, combining three different standard architectures as networks for feature extraction (backbone), and with two different networks for predicting keypoints on the heel (head networks). Using transfer learning, the backbone networks were pre-trained on the ImageNet dataset, and the combined networks (backbone + head) were fine-tuned on data from 184 trials of older, unimpaired adults. The data was recorded at three different locations and consisted of 193 k side-view images and 110 k frontal-view images. We evaluated the six different models using the absolute distance on the floor between predicted keypoints and manually labelled keypoints. For the best-performing convnet, the median error was 0.55 cm and the 75% quartile was below 1.26 cm using data from the side-view camera. The predictions are overall accurate, but show some outliers. The results indicate potential for future clinical use by automating a key step in marker-less gait parameter extraction.

PMID:39292381 | DOI:10.1007/s11517-024-03189-7

Ann Biomed Eng. 2024 Sep 18. doi: 10.1007/s10439-024-03611-z. Online ahead of print.

ABSTRACT

Computational modeling of cardiovascular function has become a critical part of diagnosing, treating and understanding cardiovascular disease. Most strategies involve constructing anatomically accurate computer models of cardiovascular structures, which is a multistep, time-consuming process. To improve the model generation process, we herein present SeqSeg (sequential segmentation): a novel deep learning-based automatic tracing and segmentation algorithm for constructing image-based vascular models. SeqSeg leverages local U-Net-based inference to sequentially segment vascular structures from medical image volumes. We tested SeqSeg on CT and MR images of aortic and aortofemoral models and compared the predictions to those of benchmark 2D and 3D global nnU-Net models, which have previously shown excellent accuracy for medical image segmentation. We demonstrate that SeqSeg is able to segment more complete vasculature and is able to generalize to vascular structures not annotated in the training data.

PMID:39292327 | DOI:10.1007/s10439-024-03611-z

Int J Clin Oncol. 2024 Sep 18. doi: 10.1007/s10147-024-02617-w. Online ahead of print.

ABSTRACT

BACKGROUND: Genome DNA methylation profiling is a promising yet costly method for cancer classification, involving substantial data. We developed an ensemble learning model to identify cancer types using methylation profiles from a limited number of CpG sites.

METHODS: Analyzing methylation data from 890 samples across 10 cancer types from the TCGA database, we utilized ANOVA and Gain Ratio to select the most significant CpG sites, then employed Gradient Boosting to reduce these to just 100 sites.

RESULTS: This approach maintained high accuracy across multiple machine learning models, with classification accuracy rates between 87.7% and 93.5% for methods including Extreme Gradient Boosting, CatBoost, and Random Forest. This method effectively minimizes the number of features needed without losing performance, helping to classify primary organs and uncover subgroups within specific cancers like breast and lung.

CONCLUSIONS: Using a gradient boosting feature selector shows potential for streamlining methylation-based cancer classification.

PMID:39292320 | DOI:10.1007/s10147-024-02617-w

Med Phys. 2024 Sep 18. doi: 10.1002/mp.17410. Online ahead of print.

ABSTRACT

BACKGROUND: The advancements in artificial intelligence and computational power have made deep learning an attractive tool for radiotherapy treatment planning. Deep learning has the potential to significantly simplify the trial-and-error process involved in inverse planning required by modern treatment techniques such as volumetric modulated arc therapy (VMAT). In this study, we explore the ability of deep learning to predict organ-at-risk (OAR) dose-volume histograms (DVHs) of left-sided breast cancer patients undergoing VMAT treatment based solely on their anatomical characteristics. The predicted DVHs could be used to derive patient-specific dose constraints and dose objectives, streamlining the treatment planning process, standardizing the quality of the plans, and personalizing the treatment planning.

PURPOSE: This study aimed to develop a deep learning-based framework for the prediction of organ-specific dose-volume histograms (DVH) based on structures delineated for left-sided breast cancer treatment.

METHODS: We used a dataset of 249 left-sided breast cancer patients treated with tangential VMAT fields. We extracted delineated structures and dose distributions for each patient and derived slice-by-slice DVHs for planning target volume (PTV) and organs-at-risk. The patients were divided into training (70%, n = 174), validation (10%, n = 24), and test (20%, n = 51) sets. Collected data were used to train a deep learning model for the prediction of the DVHs based on the delineated structures. The developed deep learning model comprised a modified DenseNet architecture followed by a recurrent neural network.

RESULTS: In the independent test set (n = 51), the point-wise differences in the slice-by-slice DVHs between the clinical and predicted DVHs were small; the mean squared errors were 3.53, 1.58, 2.28, 3.37, and 1.44 [×10-4] for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively. With the derived cumulative DVHs, the mean absolute difference ± standard deviation of mean doses between the clinical and the predicted DVH were 0.08 ± 0.04 Gy, 0.24 ± 0.22 Gy, 0.73 ± 0.46 Gy, 0.07 ± 0.06 Gy, and 0.14 ± 0.14 Gy for PTV, heart, ipsilateral lung, contralateral lung, and contralateral breast, respectively.

CONCLUSIONS: The deep learning-based approach enabled automatic and reliable prediction of the DVH based on delineated structures. The predicted DVHs could potentially serve as patient-specific clinical goals used to aid treatment planning and avoid suboptimal plans or to derive optimization objectives and constraints for automated treatment planning.

PMID:39291645 | DOI:10.1002/mp.17410

J Int Med Res. 2024 Sep;52(9):3000605241263170. doi: 10.1177/03000605241263170.

ABSTRACT

Liver vessel segmentation from routinely performed medical imaging is a useful tool for diagnosis, treatment planning and delivery, and prognosis evaluation for many diseases, particularly liver cancer. A precise representation of liver anatomy is crucial to define the extent of the disease and, when suitable, the consequent resective or ablative procedure, in order to guarantee a radical treatment without sacrificing an excessive volume of healthy liver. Once mainly performed manually, with notable cost in terms of time and human energies, vessel segmentation is currently realized through the application of artificial intelligence (AI), which has gained increased interest and development of the field. Many different AI-driven models adopted for this aim have been described and can be grouped into different categories: thresholding methods, edge- and region-based methods, model-based methods, and machine learning models. The latter includes neural network and deep learning models that now represent the principal algorithms exploited for vessel segmentation. The present narrative review describes how liver vessel segmentation can be realized through AI models, with a summary of model results in terms of accuracy, and an overview on the future progress of this topic.

PMID:39291427 | DOI:10.1177/03000605241263170