Ultrason Sonochem. 2024 Nov 12;111:107154. doi: 10.1016/j.ultsonch.2024.107154. Online ahead of print.

ABSTRACT

High-intensity focused ultrasound (HIFU) is considered as an important non-invasive way for tumor ablation in deep organs. However, accurate real-time monitoring of the temperature field within HIFU focal area remains a challenge. Although ultrasound technology, compared with other approaches, is a good choice for noninvasive and real-time monitoring on the temperature distribution, traditional ultrasonic thermometry mainly relies on the backscattered signal, which is difficult for high temperature (>50 °C) measurement. Given that artificial intelligence (AI) shows significant potential for biomedical applications, we propose an AI-powered ultrasonic thermometry using an end-to-end deep neural network termed Breath-guided Multimodal Teacher-Student (BMTS), which possesses the capability to elucidate the interaction between HIFU and complex heterogeneous biological media. It has been demonstrated experimentally that two-dimension temperature distribution within HIFU focal area in deep organ can be accurately reconstructed with an average error and a frame speed of 0.8 °C and 0.37 s, respectively. Most importantly, the maximum measurable temperature for ultrasonic technology has been successfully expanded to a record value of 67 °C. This breakthrough indicates that the development of AI-powered ultrasonic thermometry is beneficial for precise HIFU therapy planning in the future.

PMID:39549669 | DOI:10.1016/j.ultsonch.2024.107154

Radiography (Lond). 2024 Nov 14;31(1):112-117. doi: 10.1016/j.radi.2024.11.002. Online ahead of print.

ABSTRACT

INTRODUCTION: Attenuation correction (AC) is necessary for accurate assessment of radioactive distribution in single photon emission computed tomography (SPECT). The method of computed tomography-based AC (CTAC) is widely used because of its accuracy. However, patients are exposed to radiation during CT examination. The purpose of this study was to generate pseudo CT images for AC from non-AC SPECT images using deep learning and evaluate the effect of deep learning-based AC in 99mTc-labeled galactosyl human serum albumin SPECT/CT imaging.

METHODS: A cycle-consistent generative network (CycleGAN) was used to generate pseudo CT images. The test cohort consisted of each one patient with normal and abnormal liver function. SPECT images were reconstructed without AC (SPECTNC), with conventional CTAC (SPECTCTAC), and with deep learning-based AC (SPECTGAN). The accuracy of each AC was evaluated using the total liver count and the structural similarity index (SSIM) of SPECTCTAC and SPECTGAN. The coefficient of variation (%CV) was used to assess uniformity.

RESULTS: The total liver counts in SPECTGAN were significantly improved over those in SPECTNC and differed from those of SPECTCTAC by approximately 7 % in both patients. The %CV in SPECTCTAC and SPECTGAN were significantly lower than those in SPECTNC. The mean SSIM in SPECTCTAC and SPECTGAN for patients with normal and abnormal liver functions were 0.985 and 0.977, respectively.

CONCLUSIONS: The accuracy of AC with a deep learning-based method was similarly performed as the conventional CTAC. Our proposed method used only non-AC SPECT images for AC, which has great potential to reduce patient exposure by eliminating CT examination.

IMPLICATIONS FOR PRACTICE: AC of 99mTc-GSA was achieved using pseudo CT images generated with CycleGAN. Further studies on changing liver morphology and various hepatic diseases are recommended.

PMID:39549604 | DOI:10.1016/j.radi.2024.11.002

Med Biol Eng Comput. 2024 Nov 16. doi: 10.1007/s11517-024-03243-4. Online ahead of print.

ABSTRACT

Stroke, a major cause of death and disability worldwide, can be haemorrhagic or ischaemic depending on the type of bleeding in the brain. Rapid and accurate identification of stroke type and lesion segmentation is critical for timely and effective treatment. However, existing research primarily focuses on segmenting a single stroke type, potentially limiting their clinical applicability. This study addresses this gap by exploring multi-type stroke lesion segmentation using deep learning methods. Specifically, we investigate two distinct approaches: a single-stage approach that directly segments all tissue types in one model and a hierarchical approach that first classifies stroke types and then utilises specialised segmentation models for each subtype. Recognising the importance of accurate stroke classification for the hierarchical approach, we evaluate ResNet, ResNeXt and ViT networks, incorporating focal loss and oversampling techniques to mitigate the impact of class imbalance. We further explore the performance of U-Net, U-Net++ and DeepLabV3 models for segmentation within each approach. We use a comprehensive dataset of 6650 images provided by the Ministry of Health of the Republic of Türkiye. This dataset includes 1130 ischaemic strokes, 1093 haemorrhagic strokes and 4427 non-stroke cases. In our comparative experiments, we achieve an AUC score of 0.996 when classifying stroke and non-stroke slices. For lesion segmentation task, while the performance of different architectures is comparable, the hierarchical training approach outperforms the single-stage approach in terms of intersection over union (IoU). The performance of the U-Net model increased significantly from an IoU of 0.788 to 0.875 when the hierarchical approach is used. This comparative analysis aims to identify the most effective approach and deep learning model for multi-type stroke lesion segmentation in brain CT scans, potentially leading to improved clinical decision-making, treatment efficiency and outcomes.

PMID:39549224 | DOI:10.1007/s11517-024-03243-4

J Mol Evol. 2024 Nov 16. doi: 10.1007/s00239-024-10219-3. Online ahead of print.

ABSTRACT

Cold-active enzymes have recently gained popularity because of their high activity at lower temperatures than their mesophilic and thermophilic counterparts, enabling them to withstand harsh reaction conditions and enhance industrial processes. Cold-active lipases are enzymes produced by psychrophiles that live and thrive in extremely cold conditions. Cold-active lipase applications are now growing in the detergency, synthesis of fine chemicals, food processing, bioremediation, and pharmaceutical industries. The cold adaptation mechanisms exhibited by these enzymes are yet to be fully understood. Using phylogenetic analysis, and advanced deep learning-based protein structure prediction tool Alphafold2, we identified an evolutionary processes in which a conserved cold-active-like motif is presence in a distinct subclade of the tree and further predicted and simulated the three-dimensional structure of a putative cold-active lipase with the cold active motif, Glalip03, from Glaciozyma antarctica PI12. Molecular dynamics at low temperatures have revealed global stability over a wide range of temperatures, flexibility, and the ability to cope with changes in water and solvent entropy. Therefore, the knowledge we uncover here will be crucial for future research into how these low-temperature-adapted enzymes maintain their overall flexibility and function at lower temperatures.

PMID:39549052 | DOI:10.1007/s00239-024-10219-3

J Evol Biol. 2024 Nov 16:voae141. doi: 10.1093/jeb/voae141. Online ahead of print.

ABSTRACT

Amino acid substitution models play an important role in studying the evolutionary relationships among species from protein sequences. The amino acid substitution model consists of a large number of parameters; therefore, it is estimated from hundreds or thousands of alignments. Both general models and clade-specific models have been estimated and widely used in phylogenetic analyses. The maximum likelihood method is normally used to select the best fit model for a specific protein alignment under the study. A number of studies have discussed theoretical concerns as well as computational burden of the maximum likelihood methods in model selection. Recently, machine learning methods have been proposed for selecting nucleotide models. In this paper, we propose a method to measure substitution rates among amino acids (called summary statistics) from protein alignments to efficiently train a deep learning network of so-called ModelDetector for detecting amino acid substitution models. The ModelDetector network was trained from 2,246,400 alignments on a computer with 8 cores (without GPU) in about 3.3 hours. Experiments on simulation data showed that the accuracy of the ModelDetector was comparable with that of the maximum likelihood method ModelFinder. It was orders of magnitude faster than the maximum likelihood method in inferring amino acid substitution models and able to analyze genome alignments with millions of sites in minutes. The results indicate that the deep learning network can play as a promising tool for amino acid substitution model selection.

PMID:39548851 | DOI:10.1093/jeb/voae141

J Dent Res. 2024 Nov 16:220345241292566. doi: 10.1177/00220345241292566. Online ahead of print.

ABSTRACT

Accurately delineating individual teeth in 3-dimensional tooth point clouds is an important orthodontic application. Learning-based segmentation methods rely on labeled datasets, which are typically limited in scale due to the labor-intensive process of annotating each tooth. In this article, we propose a self-supervised pretraining framework, named Geo-Net, to boost segmentation performance by leveraging large-scale unlabeled data. The framework is based on the scalable masked autoencoders, and 2 geometry-guided designs, curvature-aware patching algorithm (CPA) and scale-aware reconstruction (SCR), are proposed to enhance the masked pretraining for tooth point cloud segmentation. In particular, CPA is designed to assemble informative patches as the reconstruction unit, guided by the estimated pointwise curvatures. Aimed at equipping the pretrained encoder with scale-aware modeling capacity, we also propose SCR to perform multiple reconstructions across shallow and deep layers. In vitro experiments reveal that after pretraining with large-scale unlabeled data, the proposed Geo-Net can outperform the supervised counterparts in mean Intersection of Union (mIoU) with the same amount of annotated labeled data. The code and data are available at https://github.com/yifliu3/Geo-Net.

PMID:39548729 | DOI:10.1177/00220345241292566

Protein Sci. 2024 Dec;33(12):e5211. doi: 10.1002/pro.5211.

ABSTRACT

Machine learning (ML) constructs predictive models by understanding the relationship between protein sequences and their functions, enabling efficient identification of protein sequences with high fitness values without falling into local optima, like directional evolution. However, how to extract the most pertinent functional feature information from a limited number of protein sequences is vital for optimizing the performance of ML models. Here, we propose scut_ProFP (Protein Fitness Predictor), a predictive framework that integrates feature combination and feature selection techniques. Feature combination offers comprehensive sequence information, while feature selection searches for the most beneficial features to enhance model performance, enabling accurate sequence-to-function mapping. Compared to similar frameworks, scut_ProFP demonstrates superior performance and is also competitive with more complex deep learning models-ECNet, EVmutation, and UniRep. In addition, scut_ProFP enables generalization from low-order mutants to high-order mutants. Finally, we utilized scut_ProFP to simulate the engineering of the fluorescent protein CreiLOV and highly enriched mutants with high fluorescence based on only a small number of low-fluorescence mutants. Essentially, the developed method is advantageous for ML in protein engineering, providing an effective approach to data-driven protein engineering. The code and datasets for scut_ProFP are available at https://github.com/Zhang66-star/scut_ProFP.

PMID:39548358 | DOI:10.1002/pro.5211

Sci Rep. 2024 Nov 15;14(1):28234. doi: 10.1038/s41598-024-79919-6.

ABSTRACT

The accurate detection of tunnel lining cracks and prompt identification of their primary causes are critical for maintaining tunnel availability. The advancement of deep learning, particularly in the domain of convolutional neural network (CNN) for image segmentation, has made tunnel lining crack detection more feasible. However, the CNN-based technique for tunnel lining crack detection commonly prioritizes increasing algorithmic complexity to enhance detection accuracy, posing a challenge in balancing the accuracy of detection and the efficiency of the algorithm. Motivated by the superior performance of Unet in image segmentation, this paper proposes a lightweight tunnel lining crack detection model named Mini-Unet, which refined the Unet architecture and utilized depthwise separable convolutions (DSConv) to replace some standard convolution layers. In the optimization of the proposed model parameters, applying a hybrid loss function that integrated dice loss and cross-entropy loss effectively tackled the imbalance between crack and background categories. Several models were set up to contrast with Mini-Unet and the experimental results were analyzed. Mini-Unet achieves a mean intersection over union (MIoU) of 60.76%, a mean precision of 84.18%, and a frame per second (FPS) of 5.635, respectively. Mini-Unet outperforms several mainstream models, enabling rapid detection while maintaining identified accuracy and facilitating the practical application of AI power for real-time tunnel lining crack detection.

PMID:39548331 | DOI:10.1038/s41598-024-79919-6

Sci Rep. 2024 Nov 15;14(1):28203. doi: 10.1038/s41598-024-79793-2.

ABSTRACT

This paper presents a novel approach to agricultural disease diagnostics through the integration of Deep Learning (DL) techniques with Visual Question Answering (VQA) systems, specifically targeting the detection of wheat rust. Wheat rust is a pervasive and destructive disease that significantly impacts wheat production worldwide. Traditional diagnostic methods often require expert knowledge and time-consuming processes, making rapid and accurate detection challenging. We drafted a new, WheatRustDL2024 dataset (7998 images of healthy and infected leaves) specifically designed for VQA in the context of wheat rust detection and utilized it to retrieve the initial weights on the federated learning server. This dataset comprises high-resolution images of wheat plants, annotated with detailed questions and answers pertaining to the presence, type, and severity of rust infections. Our dataset also contains images collected from various sources and successfully highlights a wide range of conditions (different lighting, obstructions in the image, etc.) in which a wheat image may be taken, therefore making a generalized universally applicable model. The trained model was federated using Flower. Following extensive analysis, the chosen central model was ResNet. Our fine-tuned ResNet achieved an accuracy of 97.69% on the existing data. We also implemented the BLIP (Bootstrapping Language-Image Pre-training) methods that enable the model to understand complex visual and textual inputs, thereby improving the accuracy and relevance of the generated answers. The dual attention mechanism, combined with BLIP techniques, allows the model to simultaneously focus on relevant image regions and pertinent parts of the questions. We also created a custom dataset (WheatRustVQA) with our augmented dataset containing 1800 augmented images and their associated question-answer pairs. The model fetches an answer with an average BLEU score of 0.6235 on our testing partition of the dataset. This federated model is lightweight and can be seamlessly integrated into mobile phones, drones, etc. without any hardware requirement. Our results indicate that integrating deep learning with VQA for agricultural disease diagnostics not only accelerates the detection process but also reduces dependency on human experts, making it a valuable tool for farmers and agricultural professionals. This approach holds promise for broader applications in plant pathology and precision agriculture and can consequently address food security issues.

PMID:39548249 | DOI:10.1038/s41598-024-79793-2

Sci Rep. 2024 Nov 15;14(1):28246. doi: 10.1038/s41598-024-79249-7.

ABSTRACT

Given a set of labels, multi-label text classification (MLTC) aims to assign multiple relevant labels for a text. Recently, deep learning models get inspiring results in MLTC. Training a high-quality deep MLTC model typically demands large-scale labeled data. And comparing with annotations for single-label data samples, annotations for multi-label samples are typically more time-consuming and expensive. Active learning can enable a classification model to achieve optimal prediction performance using fewer labeled samples. Although active learning has been considered for deep learning models, there are few studies on active learning for deep multi-label classification models. In this work, for the deep MLTC model, we propose a deep Active Learning method based on Bayesian deep learning and Expected confidence (BEAL). It adopts Bayesian deep learning to derive the deep model's posterior predictive distribution and defines a new expected confidence-based acquisition function to select uncertain samples for deep MLTC model training. Moreover, we perform experiments with a BERT-based MLTC model, where BERT can achieve satisfactory performance by fine-tuning in various classification tasks. The results on benchmark datasets demonstrate that BEAL enables more efficient model training, allowing the deep model to achieve training convergence with fewer labeled samples.

PMID:39548182 | DOI:10.1038/s41598-024-79249-7

Sci Rep. 2024 Nov 15;14(1):28224. doi: 10.1038/s41598-024-76569-6.

ABSTRACT

Numerous Deep Learning (DL) scenarios have been developed for evolving new healthcare systems that leverage large datasets, distributed computing, and the Internet of Things (IoT). However, the data used in these scenarios tend to be noisy, necessitating the incorporation of robust pre-processing techniques, including data cleaning, preparation, normalization, and addressing imbalances. These steps are crucial for generating a robust dataset for training. Designing frameworks capable of handling such data without compromising efficiency is essential to ensuring robustness. This research aims to propose a novel healthcare framework that selects the best features and enhances performance. This robust deep learning framework, called (R-DLH2O), is designed for forecasting cardiovascular crises. Unlike existing methods, R-DLH2O integrates five distinct phases: robust pre-processing, feature selection, feed-forward neural network, prediction, and performance evaluation. This multi-phase approach ensures superior accuracy and efficiency in crisis prediction, offering a significant advancement in healthcare analytics. H2O is utilized in the R-DLH2O framework for processing big data. The main improvement of this paper lies in the unique form of the Whale Optimization Algorithm (WOA), specifically the Modified WOA (MWOA). The Gaussian distribution approach for random walks was employed with the diffusion strategy to choose the optimal MWOA solution during the growth phase. To validate the R-DLH2O framework, six performance tests were conducted. Surprisingly, the MWOA-2 outperformed other heuristic algorithms in speed, despite exhibiting lower accuracy and scalability. The suggested MWOA was further analyzed using benchmark functions from CEC2005, demonstrating its advantages in accuracy and robustness over WOA. These findings highlight that the framework's processing time is 436 s, mean per-class error is 0.150125, accuracy 95.93%, precision 92.57%, and recall 93.6% across all datasets. These findings highlight the framework's potential to produce significant and robust results, outperforming previous frameworks concerning time and accuracy.

PMID:39548142 | DOI:10.1038/s41598-024-76569-6

Sci Rep. 2024 Nov 15;14(1):28119. doi: 10.1038/s41598-024-72580-z.

ABSTRACT

China's operational highway subgrades exhibit a trend of diversifying types and an increasing number of defects, leading to more frequent urban road safety incidents. This paper starts from the non-destructive testing of urban road subgrade defects using geological radar, aiming to achieve intelligent identification of subgrade pathologies with geological radar. The GprMax forward simulation software is used to establish multi-layer composite structural models of the subgrade, studying the characteristics of geological radar images for different types of subgrade diseases. Based on the forward simulation images of geological radar for subgrade defects and field measurement data, a geological radar subgrade defect image database is established. The Faster R-CNN deep learning algorithm is applied to achieve target detection, recognition, and classification of subgrade defect images. By comparing the loss value, total number of identified regions, and recognition accuracy as metrics, the study compares four improved versions of the Faster R-CNN algorithm. The results indicate that the faster_rcnn_inception_v2 version is more suitable for the intelligent identification of non-destructive testing of urban road subgrade defects.

PMID:39548115 | DOI:10.1038/s41598-024-72580-z

Sci Rep. 2024 Nov 15;14(1):28165. doi: 10.1038/s41598-024-72522-9.

ABSTRACT

To examine the morphological impact of deep learning (DL)-quantified biomarkers on point-wise sensitivity (PWS) using microperimetry (MP) and optical coherence tomography (OCT) in intermediate AMD (iAMD). Patients with iAMD were examined by OCT (Spectralis). DL-based algorithms quantified ellipsoid zone (EZ)-thickness, hyperreflective foci (HRF) and drusen volume. Outer nuclear layer (ONL)-thickness and subretinal drusenoid deposits (SDD) were quantified by human experts. All patients completed four MP examinations using an identical custom 45 stimuli grid on MP-3 (NIDEK) and MAIA (CenterVue). MP stimuli were co-registered with corresponding OCT using image registration algorithms. Multivariable mixed-effect models were calculated. 3.600 PWS from 20 eyes of 20 patients were analyzed. Decreased EZ thickness, decreased ONL thickness, increased HRF and increased drusen volume had a significant negative effect on PWS (all p < 0.001) with significant interaction with eccentricity (p < 0.001). Mean PWS was 26.25 ± 3.43 dB on MP3 and 22.63 ± 3.69 dB on MAIA. Univariate analyses revealed a negative association of PWS and SDD (p < 0.001). Subclinical changes in EZ integrity, HRF and drusen volume are quantifiable structural biomarkers associated with reduced retinal function. Topographic co-registration between structure on OCT volumes and sensitivity in MP broadens the understanding of pathognomonic biomarkers with potential for evaluation of quantifiable functional endpoints.

PMID:39548108 | DOI:10.1038/s41598-024-72522-9

Sci Data. 2024 Nov 15;11(1):1233. doi: 10.1038/s41597-024-04041-1.

ABSTRACT

Blood pressure (BP) is one of the most prominent indicators of potential cardiovascular disorders. Traditionally, BP measurement relies on inflatable cuffs, which is inconvenient and limit the acquisition of such important health-related information in general population. Based on large amounts of well-collected and annotated data, deep-learning approaches present a generalization potential that arose as an alternative to enable more pervasive approaches. However, most existing work in this area currently uses datasets with limitations, such as lack of subject identification and severe data imbalance that can result in data leakage and algorithm bias. Thus, to offer a more properly curated source of information, we propose a derivative dataset composed of 380 hours of the most common biomedical signals, including arterial blood pressure, photoplethysmography, and electrocardiogram for 1,524 anonymized subjects, each having 30 segments of 30 seconds of those signals. We also validated the proposed dataset through experiments using state-of-the-art deep-learning methods, as we highlight the importance of standardized benchmarks for calibration-free blood pressure estimation scenarios.

PMID:39548096 | DOI:10.1038/s41597-024-04041-1

J Cancer Res Clin Oncol. 2024 Nov 16;150(11):504. doi: 10.1007/s00432-024-06027-3.

NO ABSTRACT

PMID:39547976 | DOI:10.1007/s00432-024-06027-3

Eur Urol Oncol. 2024 Nov 14:S2588-9311(24)00248-7. doi: 10.1016/j.euo.2024.11.001. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVE: Magnetic resonance imaging (MRI) plays a critical role in prostate cancer diagnosis, but is limited by variability in interpretation and diagnostic accuracy. This systematic review evaluates the current state of deep learning (DL) models in enhancing the automatic detection, localization, and characterization of clinically significant prostate cancer (csPCa) on MRI.

METHODS: A systematic search was conducted across Medline/PubMed, Embase, Web of Science, and ScienceDirect for studies published between January 2020 and September 2023. Studies were included if these presented and validated fully automated DL models for csPCa detection on MRI, with pathology confirmation. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool and the Checklist for Artificial Intelligence in Medical Imaging.

KEY FINDINGS AND LIMITATIONS: Twenty-five studies met the inclusion criteria, showing promising results in detecting and characterizing csPCa. However, significant heterogeneity in study designs, validation strategies, and datasets complicates direct comparisons. Only one-third of studies performed external validation, highlighting a critical gap in generalizability. The reliance on internal validation limits a broader application of these findings, and the lack of standardized methodologies hinders the integration of DL models into clinical practice.

CONCLUSIONS AND CLINICAL IMPLICATIONS: DL models demonstrate significant potential in improving prostate cancer diagnostics on MRI. However, challenges in validation, generalizability, and clinical implementation must be addressed. Future research should focus on standardizing methodologies, ensuring external validation and conducting prospective clinical trials to facilitate the adoption of artificial intelligence (AI) in routine clinical settings. These findings support the cautious integration of AI into clinical practice, with further studies needed to confirm their efficacy in diverse clinical environments.

PATIENT SUMMARY: In this study, we reviewed how artificial intelligence (AI) models can help doctors better detect and understand aggressive prostate cancer using magnetic resonance imaging scans. We found that while these AI tools show promise, these tools need more testing and validation in different hospitals before these can be used widely in patient care.

PMID:39547898 | DOI:10.1016/j.euo.2024.11.001

J Med Internet Res. 2024 Nov 15;26:e60057. doi: 10.2196/60057.

ABSTRACT

BACKGROUND: As the digital health landscape continues to evolve, analyzing the progress and direction of the field can yield valuable insights. The Journal of Medical Internet Research (JMIR) has been at the forefront of disseminating digital health research since 1999. A comprehensive network analysis of JMIR publications can help illuminate the evolution and trends in digital medicine over the past 25 years.

OBJECTIVE: This study aims to conduct a detailed network analysis of JMIR's publications to uncover the growth patterns, dominant themes, and potential future trajectories in digital health research.

METHODS: We retrieved 8068 JMIR papers from PubMed using the Biopython library. Keyword metrics were assessed using accuracy, recall, and F1-scores to evaluate the effectiveness of keyword identification from Claude 3 Opus and Gemini 1.5 Pro in addition to 2 conventional natural language processing methods using key bidirectional encoder representations from transformers. Future trends for 2024-2026 were predicted using Claude 3 Opus, Google's Time Series Foundation Model, autoregressive integrated moving average, exponential smoothing, and Prophet. Network visualization techniques were used to represent and analyze the complex relationships between collaborating countries, paper types, and keyword co-occurrence.

RESULTS: JMIR's publication volume showed consistent growth, with a peak in 2020. The United States dominated country contributions, with China showing a notable increase in recent years. Keyword analysis from 1999 to 2023 showed significant thematic shifts, from an early internet and digital health focus to the dominance of COVID-19 and advanced technologies such as machine learning. Predictions for 2024-2026 suggest an increased focus on artificial intelligence, digital health, and mental health.

CONCLUSIONS: Network analysis of JMIR publications provides a macroscopic view of the evolution of the digital health field. The journal's trajectory reflects broader technological advances and shifting research priorities, including the impact of the COVID-19 pandemic. The predicted trends underscore the growing importance of computational technology in future health care research and practice. The findings from JMIR provide a glimpse into the future of digital medicine, suggesting a robust integration of artificial intelligence and continued emphasis on mental health in the postpandemic era.

PMID:39546778 | DOI:10.2196/60057

J Med Internet Res. 2024 Nov 15;26:e51432. doi: 10.2196/51432.

ABSTRACT

BACKGROUND: To accurately capture an individual's food intake, dietitians are often required to ask clients about their food frequencies and portions, and they have to rely on the client's memory, which can be burdensome. While taking food photos alongside food records can alleviate user burden and reduce errors in self-reporting, this method still requires trained staff to translate food photos into dietary intake data. Image-assisted dietary assessment (IADA) is an innovative approach that uses computer algorithms to mimic human performance in estimating dietary information from food images. This field has seen continuous improvement through advancements in computer science, particularly in artificial intelligence (AI). However, the technical nature of this field can make it challenging for those without a technical background to understand it completely.

OBJECTIVE: This review aims to fill the gap by providing a current overview of AI's integration into dietary assessment using food images. The content is organized chronologically and presented in an accessible manner for those unfamiliar with AI terminology. In addition, we discuss the systems' strengths and weaknesses and propose enhancements to improve IADA's accuracy and adoption in the nutrition community.

METHODS: This scoping review used PubMed and Google Scholar databases to identify relevant studies. The review focused on computational techniques used in IADA, specifically AI models, devices, and sensors, or digital methods for food recognition and food volume estimation published between 2008 and 2021.

RESULTS: A total of 522 articles were initially identified. On the basis of a rigorous selection process, 84 (16.1%) articles were ultimately included in this review. The selected articles reveal that early systems, developed before 2015, relied on handcrafted machine learning algorithms to manage traditional sequential processes, such as segmentation, food identification, portion estimation, and nutrient calculations. Since 2015, these handcrafted algorithms have been largely replaced by deep learning algorithms for handling the same tasks. More recently, the traditional sequential process has been superseded by advanced algorithms, including multitask convolutional neural networks and generative adversarial networks. Most of the systems were validated for macronutrient and energy estimation, while only a few were capable of estimating micronutrients, such as sodium. Notably, significant advancements have been made in the field of IADA, with efforts focused on replicating humanlike performance.

CONCLUSIONS: This review highlights the progress made by IADA, particularly in the areas of food identification and portion estimation. Advancements in AI techniques have shown great potential to improve the accuracy and efficiency of this field. However, it is crucial to involve dietitians and nutritionists in the development of these systems to ensure they meet the requirements and trust of professionals in the field.

PMID:39546777 | DOI:10.2196/51432

Langmuir. 2024 Nov 15. doi: 10.1021/acs.langmuir.4c02867. Online ahead of print.

ABSTRACT

Nanostructured particles (NSPs), with their remarkable properties at the nanoscale, possess key functions required for unlocking a sustainable future. Fabricating these particles using aerosol methods and spraying processes enables precise control over the particle morphology, structure, composition, and crystallinity during in-flight transformation. In this Perspective, the significant impact of NSPs on technological advancement for energy and environmental applications is discussed. Furthermore, incorporating in situ/operando assessment techniques alongside machine and deep learning is explored. Finally, the future development trends and the perspective on the advancing NSPs synthesis via aerosol process are elaborated for further driving innovations for supersmart and carbon-neutral society.

PMID:39546762 | DOI:10.1021/acs.langmuir.4c02867

Opt Lett. 2024 Nov 15;49(22):6457-6460. doi: 10.1364/OL.533923.

ABSTRACT

The acquisition of the time of flight (ToF) of photons has found numerous applications in the biomedical field. Over the last decades, a few strategies have been proposed to deconvolve the temporal instrument response function (IRF) that distorts the experimental time-resolved data. However, these methods require burdensome computational strategies and regularization terms to mitigate noise contributions. Herein, we propose a deep learning model specifically to perform the deconvolution task in fluorescence lifetime imaging (FLI). The model is trained and validated with representative simulated FLI data with the goal of retrieving the true photon ToF distribution. Its performance and robustness are validated with well-controlled in vitro experiments using three time-resolved imaging modalities with markedly different temporal IRFs. The model aptitude is further established with in vivo preclinical investigation. Overall, these in vitro and in vivo validations demonstrate the flexibility and accuracy of deep learning model-based deconvolution in time-resolved FLI and diffuse optical imaging.

PMID:39546693 | DOI:10.1364/OL.533923