Eur Radiol. 2025 Feb 4. doi: 10.1007/s00330-025-11386-7. Online ahead of print.

ABSTRACT

OBJECTIVES: Thyroid cancer, the only cancer that uses age as a specific predictor of survival, is increasing in incidence, yet it has a low mortality rate, which can lead to overdiagnosis and overtreatment. We developed an age-stratified deep learning (DL) model (hereafter, ASMCNet) for classifying thyroid nodules and aimed to investigate the effect of age stratification on the accuracy of a DL model, exploring how ASMCNet can help radiologists improve diagnostic performance and avoid unnecessary biopsies.

METHODS: In this retrospective study, we used ultrasound images from three hospitals, a total of 10,391 images of 5934 patients were used for training, validation, and testing. The performance of ASMCNet was compared with that of model-trained non-age-stratified radiologists with different experience levels on the test data set with the DeLong method.

RESULTS: The area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity of ASMCNet were 0.906, 86.1%, and 85.1%, respectively, which exceeded those of model-trained non-age-stratified (0.867, 83.2%, and 75.5%, respectively; p < 0.001) and higher than all of the radiologists (p < 0.001). Reader studies show that radiologists' performances are improved when assisted by the explaining heatmaps (p < 0.001).

CONCLUSIONS: Our study demonstrates that age stratification based on DL can further improve the performance of thyroid tumor classification models, which also suggests that age is an important factor in the diagnosis of thyroid tumors. The ASMCNet model shows promising clinical applicability and can assist radiologists in improving diagnostic accuracy.

KEY POINTS: Question Age is crucial for differentiated thyroid carcinoma (DTC) prognosis, yet its diagnostic impact lacks research. Findings Adding age stratification to DL models can further improve the accuracy of thyroid nodule diagnosis. Clinical relevance Age-stratified multimodal classification network is a reliable tool used to help radiologists diagnose thyroid nodules, and integrating it into clinical practice can improve diagnostic accuracy and reduce unnecessary biopsies or treatments.

PMID:39903238 | DOI:10.1007/s00330-025-11386-7

Invest Ophthalmol Vis Sci. 2025 Feb 3;66(2):6. doi: 10.1167/iovs.66.2.6.

ABSTRACT

PURPOSE: The purpose of this study was to evaluate the feasibility and utility of optical coherence tomography (OCT)-based, targeted microperimetry grids in assessing focal lesions in intermediate age-related macular degeneration (iAMD).

METHODS: The multicenter, prospective PINNACLE study enrolled 395 patients with iAMD aged 55 to 90 years across 12 international sites. Participants underwent imaging, including OCT and microperimetry, every 4 to 12 months over 3 years. Deep learning algorithms detected focal lesions and changes in OCT images, including drusen regression, EZ/IZ loss with hypertransmission, and subretinal fluid, guiding 5-point microperimetry targeted to lesion locations. Data were analyzed using linear mixed models to estimate differences between retinal sensitivity measured by the 5-point focal grids and sensitivity interpolated from the 24-point standard grids.

RESULTS: The final analysis included 93 eyes from 83 patients, assessing 605 of the 5-point targeted grids and standard grids across 235 focal lesions. The Pearson correlation between focally measured sensitivity and interpolated sensitivity was 0.76. However, interpolation from the standard grid could be erroneous, especially in central regions of lesions characterized by EZ/IZ loss with hypertransmission and subretinal fluid. Interpolation errors increased with distance to the nearest measurement point (slope = 2.20 dB per degree, 95% confidence interval [CI] = 1.52 to 2.87). A significant negative relationship was found between interpolation errors and retinal sensitivity, with the highest errors in areas of low sensitivity. Lesion size significantly impacted interpolation errors for EZ/IZ loss with hypertransmission (slope = -19.41 dB/mm², 95% CI = -29.63 to -9.18).

CONCLUSIONS: Targeted grids improved the detection and understanding of how focal retinal changes affect visual function in patients with iAMD, supporting the development of therapeutic interventions.

PMID:39903180 | DOI:10.1167/iovs.66.2.6

Radiology. 2025 Feb;314(2):e241001. doi: 10.1148/radiol.241001.

ABSTRACT

Background Prognostic value of radiologic features in interstitial lung disease (ILD) has been predominantly studied in idiopathic pulmonary fibrosis, but findings vary. The relative importance of features versus guideline-defined patterns in predicting outcomes is unknown. Purpose To identify radiologic features that are independently associated with transplant-free survival beyond clinical predictive factors across all ILD subtypes, and to identify whether individual features versus patterns are more important for prognostication. Materials and Methods This is a secondary analysis of the prospective Canadian Registry for Pulmonary Fibrosis. Consecutive patients with ILD were evaluated in standardized multidisciplinary discussions between January 2021 and March 2022. Radiologic features on thin-section CT images were quantified, and guideline-defined usual interstitial pneumonia (UIP) and fibrotic hypersensitivity pneumonitis (fHP) patterns were assigned. Multivariable Cox analysis was used to assess the associations of radiologic features with transplant-free survival, and nested models were used to test the relative importance of features compared with patterns. Results A total of 1593 patients (mean age, 66 years ± 12 [SD]; 800 male) were included. The following four features were associated with transplant-free survival: extent of honeycombing (hazard ratio, 1.20; 95% CI; 1.06, 1.36 per 10% increase in lung involvement; P = .005), extent of traction bronchiectasis (hazard ratio, 1.18; 95% CI: 1.10, 1.26 per 10% increase; P < .001), pulmonary artery diameter (hazard ratio, 1.03; 95% CI: 1.01; 1.04 per 1-mm increase; P = .002), and presence of subpleural sparing (hazard ratio, 0.76; 95% CI: 0.56, 0.96; P = .03). Guideline-defined patterns were not independently associated with survival in a model that included these four radiologic features, each of which retained its prognostic value. Conclusion The extent of fibrosis was predictive of worse outcomes across all ILD subtypes in a dose-dependent fashion and independent of well-recognized clinical prognostic factors. Guideline-defined UIP and fHP patterns each helped risk-stratify patients in isolation but lost prognostic value when accounting for the extent of fibrosis, suggesting that their previous association with mortality is based on these patterns acting as surrogates for a greater extent of fibrosis. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Wells in this issue.

PMID:39903073 | DOI:10.1148/radiol.241001

PLoS Pathog. 2025 Feb 4;21(2):e1012813. doi: 10.1371/journal.ppat.1012813. eCollection 2025 Feb.

ABSTRACT

The var multigene family encodes Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1), central to host-parasite interactions. Genome structure studies have identified three major groups of var genes by specific upstream sequences (upsA, B, or C). Var with these ups groups have different chromosomal locations, transcriptional directions, and associations with disease severity. Here we explore temporal and spatial diversity of a region of var genes encoding the DBLα domain of PfEMP1 in Africa. By applying a novel ups classification algorithm (cUps) to publicly-available DBLα sequence datasets, we categorised DBLα according to association with the three ups groups, thereby avoiding the need to sequence complete genes. Data from deep sequencing of DBLα types in a local population in northern Ghana surveyed seven times from 2012 to 2017 found variants with rare-to-moderate-to-extreme frequencies, and the common variants were temporally stable in this local endemic area. Furthermore, we observed that every isolate repertoire, whether mono- or multiclonal, comprised DBLα types occurring with these frequency ranges implying a common genome structure. When comparing African countries of Ghana, Gabon, Malawi, and Uganda, we report that some DBLα types were consistently found at high frequencies in multiple African countries while others were common only at the country level. The implication of these local and pan-Africa population patterns is discussed in terms of advantage to the parasite with regards to within-host adaptation and resilience to malaria control.

PMID:39903780 | DOI:10.1371/journal.ppat.1012813

Melanoma Res. 2025 Feb 4. doi: 10.1097/CMR.0000000000001025. Online ahead of print.

ABSTRACT

While the association between chronic lymphocytic leukemia (CLL) and a higher incidence of melanoma is well documented, the diagnosis of concurrent high-risk chronic myelomonocytic leukemia (CMML) and metastatic melanoma (MM) has not previously been described. Moreover, the treatment of MM and CMML differ greatly in the mechanism of action of their corresponding antineoplastic therapies: treatment of MM frequently involves immune checkpoint inhibitors (ICI), while patients with CMML receive myelosuppressive agents. Simultaneous management of these malignancies can be nuanced due to the potential impact of one treatment's constituents on the activity of the other and the broad and nonoverlapping array of potential adverse effects of these agents. Here, we describe the clinical course of a patient who was diagnosed with concurrent MM and CMML and our approach to the challenging balance of delivering ICI concurrently with the hypomethylating agent azacitidine and the BCL-2 inhibitor venetoclax.

PMID:39903257 | DOI:10.1097/CMR.0000000000001025

J Virol. 2025 Feb 4:e0141324. doi: 10.1128/jvi.01413-24. Online ahead of print.

ABSTRACT

Type III interferons (IFNs) primarily act on epithelial cells and protect against virus infection of the mucosa, whereas type I IFNs act more systemically. To date, it has been unknown which epithelial subtypes in the upper airways, the primary site for initial infection for most respiratory viruses, primarily rely on type III IFN or type I IFNs for antiviral protection. To address this question, we performed a single-cell transcriptomics analysis of the epithelial IFN-mediated response focusing on the upper airways of mice. This work identified nine distinct cell types derived from the olfactory epithelium and thirteen distinct cell types from the respiratory epithelium. Interestingly, type I IFNs induced a stronger antiviral transcriptional response than type III IFN in respiratory epithelial cells, whereas in olfactory epithelial cells, including sustentacular (SUS) and Bowman's gland cells (BGC), type III IFN was more dominant compared to type I IFN. SUS and BGC, which provide structural support and maintain the integrity of olfactory sensory neurons, were highly susceptible to infection with a mouse-adapted variant of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2 MA20) but were protected against infection if the animals were prophylactically treated with type III IFN. These findings demonstrate a high degree of cell type heterogeneity in terms of interferon-mediated antiviral responses and reveal a potent role for type III IFNs in protecting the olfactory epithelium.IMPORTANCESARS-CoV-2 infects SUS and BGC in the olfactory epithelium, causing an impairment of structural support and integrity of olfactory sensory neurons that can result in severe olfactory dysfunctions. We observed an unexpected compartmentalization of the IFN-mediated transcriptional response within the airway epithelium, and we found that olfactory epithelial cells preferentially respond to type III IFN, which resulted in robust antiviral protection of SUS and BGC. Given the proximity of the olfactory epithelium to the central nervous system, we hypothesize that evolution favored a type III IFN-biased antiviral immune response in this tissue to limit inflammatory responses in the brain. Cell type-specific antiviral responses in the upper airways, triggered by the different types of IFNs, should be investigated in more detail and carefully taken into consideration during the development of IFN-based antivirals for clinical use.

PMID:39902863 | DOI:10.1128/jvi.01413-24

PLoS One. 2025 Feb 4;20(2):e0317660. doi: 10.1371/journal.pone.0317660. eCollection 2025.

ABSTRACT

BACKGROUND: Identification of real-time adverse drug reactions [ADRs] (as opposed to the risk of ADRs) in older poly-medicated people in primary care is a challenging task, often undertaken without an explicit strategy. This systematic review aims to evaluate replicable instruments and methods for identifying and addressing ADRs.

METHODS: A systematic search was conducted in Medline, CINAHL, Scopus, Web of Science and Cochrane library, using controlled vocabulary (MeSH) and free-text terms. Randomised controlled trials (RCTs) implementing strategies to identify or resolve ADRs experienced by patients in primary care were included. Two reviewers independently screened studies, extracted data, and assessed the risk of bias using the Cochrane Risk of Bias tool. Discrepancies were resolved by discussion.

RESULTS: From 2,182 unique records, 49 studies were identified for full review. Eight papers reporting results from 6 RCTs were included. All six trials utilised a list of medicine-related unwanted symptoms to identify ADRs. Two of three studies using adverse drug reaction questionnaires reported statistically significant increased rates of ADR reporting. Two of three studies that combined symptom questionnaires with prescriber consultations reported reductions in the number of health problems. Overall, results suggest that the three studies that described multidisciplinary collaborations using lists of ADRs plus prescriber reviews enhanced patient safety. However, the RCTs were unblinded and reported suboptimal retention. When considered as a whole, findings are equivocal and the data are too heterogenous to warrant any firm conclusions, beyond the need for more research to optimise strategies to safeguard patient wellbeing.

IMPLICATIONS: Adaptable and scalable instruments with decision support are needed in primary care to identify and mitigate medicine-related harm in older poly-medicated people. The effectiveness of adverse drug reaction identification instruments, the value of comprehensive instruments, and the optimum method of delivery should be explored in multicentre trials.

PMID:39903764 | DOI:10.1371/journal.pone.0317660

Clin Cancer Res. 2025 Feb 4. doi: 10.1158/1078-0432.CCR-24-1525. Online ahead of print.

ABSTRACT

Sacituzumab govitecan (SG), a Trop-2-directed antibody-drug conjugate, is currently approved to treat metastatic triple-negative breast cancer and HR+/HER2- breast cancer, and is under clinical investigation for a range of other tumor types. This review describes its mode of action, development, and clinical outcomes. SG is composed of SN-38 (a topoisomerase I inhibitor derived from irinotecan) covalently linked to an anti-Trop-2 monoclonal antibody (sacituzumab; hRS7) via a hydrolysable CL2A linker. SN-38 was chosen due to its potent antitumor activity; CL2A occupies the most effective position on SN-38 for maintaining stability during transport, with pH-sensitive payload release in the tumor, and the antigen target (Trop-2) is highly expressed on many solid tumors. SG has an ~8:1 drug-to-antibody ratio and delivers therapeutic SN-38 concentration to Trop-2+ expressing tumor cells via rapid internalization and efficient payload release. Free SN-38 can subsequently enter the tumor microenvironment and kill adjacent tumor cells with or without Trop-2 expression (bystander effect). SN-38 induces DNA breakage and inhibits nucleic acid synthesis via a drug-induced topoisomerase 1:DNA complex that interferes with cell proliferation, causing apoptosis. Dose-finding studies support SG 10 mg/kg on days 1 and 8 of a 21-day cycle as the monotherapy dose for clinical use; this was determined by therapeutic index improvement based on efficacy and safety. Payload-linker dynamics and SG potency ensure continued tissue penetration. Neutropenia and diarrhea are the most common grade ≥ 3 treatment-emergent adverse events with SG, but they are manageable. Efficacy of SG has been demonstrated across a broad spectrum of solid tumors.

PMID:39903492 | DOI:10.1158/1078-0432.CCR-24-1525

IEEE Access. 2024;12:180792-180814. doi: 10.1109/access.2024.3505038. Epub 2024 Nov 22.

ABSTRACT

Cognitive impairment and dementia-related diseases develop several years before moderate or severe deterioration in cognitive function occurs. Nevertheless, most dementia cases, especially in low- and middle-income countries, remain undiagnosed because of limited access to affordable diagnostic tools. Additionally, the development of accessible tools for diagnosing and predicting cognitive impairment has not been extensively discussed in the literature. The objective of this study is to develop a cost-effective and highly accessible machine learning model to predict the risk of cognitive impairment for up to five years before clinical insight. We utilized easily accessible data from the National Alzheimer's Coordinating Center (NACC) Uniform Data Set (UDS) to train and evaluate various machine learning and deep learning models. A novel algorithm was developed to facilitate the selection of cost-effective models that offer high performance while minimizing development and operational costs. We conducted various assessments, including feature selection, time-series analyses, and external validation of the selected model. Our findings indicated that the Support Vector Machine (SVM) model was preferred over other high-performing neural network models because of its computational efficiency, achieving F2-scores of 0.828 in cross-validation and 0.750 in a generalizability test. Additionally, we found that demographic and historical health data are valuable for early prediction of cognitive impairment. This study demonstrates the potential of developing accessible solutions to predict cognitive impairment early using accurate and efficient machine learning models. Future interventions should consider creating cost-effective assessment tools to support global action plans and reduce the risk of cognitive impairment.

PMID:39902153 | PMC:PMC11790289 | DOI:10.1109/access.2024.3505038

Pharmgenomics Pers Med. 2025 Jan 30;18:19-34. doi: 10.2147/PGPM.S490135. eCollection 2025.

ABSTRACT

Pharmacogenomics is the integration of genomics and pharmacology to optimize drug response and reduce side effects. In terms of personalized or individualized medicine, PGx is defined as the identification and analysis of specific genetic variants associated with particular drug treatments for each patient. Under a precision public health (PPH) approach, population-level data are analyzed to generate public health strategies. The objective of this study was to conduct a scoping review of technological tools, examining their evolution, the predominance of high-income countries in their development, and the gaps and needs for genomic data and advances in low- and middle-income countries (LMICs). This review was conducted in accordance with the ScPRISMA guidelines. A search was conducted in PubMed, Web of Science and Embase until January 2024. A total of 40 documents were selected, which revealed the continuous evolution and progressive development of pharmacogenomic tools. The technological tools developed come from high-income countries, particularly the United States, Canada, China, and several European nations, where international collaboration has been essential to maintain and expand these tools, which have evolved to keep pace with the rapid generation of genomic data. This trend shows a scarce development of technological tools for public health precision in LMICs, which evidences the need to increase investment in genomic research infrastructure in this aspect and in the development of capacities to guarantee global accessibility and boost PPH for all populations.

PMID:39902237 | PMC:PMC11789506 | DOI:10.2147/PGPM.S490135

Curr Cardiol Rev. 2025 Jan 31. doi: 10.2174/011573403X334668241227074314. Online ahead of print.

ABSTRACT

Pharmacogenomics has transformed the way we approach the treatment of the most common diseases worldwide, especially cardiovascular. In this article, we highlight the main categories of drugs involved in major cardiovascular diseases (CVD), related genetic variability and their effects on metabolism in each case of contrastive operability. This not only explains disparities in treatment outcomes but also unfolds customised management based on genomic studies to improve efficiency and limit side effects. Genetic variations have been identified that impact the efficacy, safety, and adverse effects of drugs commonly used in the treatment of CVDs, such as Angiotensin converting Enzyme Inhibitor (ACEI), Angiotensin Receptor Blocker (ARBs), calcium channel blockers, antiplatelet agents, diuretics, statins, beta-blockers, and anticoagulants. It discusses the impact of genetic polymorphisms on drug metabolism, efficacy, and adverse reactions, highlighting the importance of genetic testing in optimizing treatment outcomes. Pharmacogenomics holds immense potential for revolutionizing the management of CVDs by enabling personalized medicine approaches tailored to individual genetic profiles. However, challenges such as clinical implementation, cost-effectiveness, and ethical considerations need to be addressed to completely incorporate pharmacogenomic testing into standard clinical practice. Continued research and clinical diligence are required for the utilization of pharmacogenomics to improve therapeutic outcomes and reduce the burden of CVD globally.

PMID:39901689 | DOI:10.2174/011573403X334668241227074314

Pharmacol Res Perspect. 2025 Feb;13(1):e70065. doi: 10.1002/prp2.70065.

ABSTRACT

Pharmacogenomics is a field of personalized medicine that aims to tailor drug dosing based on the genetics of an individual. The polymorphic and complex CYP2D6 gene is important to analyze because of its role in the metabolism of approximately a quarter of all drugs. Several bioinformatic tools have been developed to genotype CYP2D6 from short-read sequencing data. Among these, Cyrius, a tool specifically designed for CYP2D6 genotyping, has demonstrated high performance across various datasets. However, Cyrius has not been updated in the past 3 years, during which dozens of new star alleles have been identified and some previously defined ones revised. In this work, we simulated all known CYP2D6 haplotypes to assess the ability of Cyrius to identify them. In that dataset, Cyrius was unable to call or misidentified 50 of 360 samples. Given the importance of providing an up-to-date tool, particularly in clinical settings, we present an upgraded version of the tool, named BCyrius, which includes all the missing star alleles as well as revisions to the previously listed ones. BCyrius successfully identified 100% of the currently defined minor star alleles, higher than Cyrius (85.6%) and the two other tested tools, Aldy and StellarPGx, which identified 92.2% and 87.8%, respectively. BCyrius also demonstrated slightly improved performance on a dataset of real biological samples, resulting in a higher call rate while maintaining similar accuracy with Cyrius. In addition to providing genotyping results, BCyrius also reports the predicted phenotype, along with information for each detected haplotype, including population frequencies.

PMID:39901590 | DOI:10.1002/prp2.70065

Proc Inst Mech Eng H. 2025 Feb 4:9544119251315877. doi: 10.1177/09544119251315877. Online ahead of print.

ABSTRACT

Generative deep learning has emerged as a promising data augmentation technique in recent years. This approach becomes particularly valuable in areas such as motion analysis, where it is challenging to collect substantial amounts of data. The objective of the current study is to introduce a data augmentation strategy that relies on a variational autoencoder to generate synthetic data of kinetic and kinematic variables. The kinematic and kinetic variables consist of hip and knee joint angles and moments, respectively, in both sagittal and frontal plane, and ground reaction forces. Statistical parametric mapping (SPM) did not detect significant differences between real and synthetic data for each of the biomechanical variables considered. To further evaluate the effectiveness of this approach, a long-short term model (LSTM) was trained both only on real data (R) and on the combination of real and synthetic data (R&S); the performance of each of these two trained models was then assessed on real test data unseen during training. The principal findings included achieving comparable results in terms of nRMSE when predicting knee joint moments in the frontal (R&S: 9.86% vs R: 10.72%) and sagittal plane (R&S: 9.21% vs R: 9.75%), and hip joint moments in the frontal (R&S: 16.93% vs R: 16.79%) and sagittal plane (R&S: 13.29% vs R: 14.60%). The main novelty of this study lies in introducing an effective data augmentation approach in motion analysis settings.

PMID:39902572 | DOI:10.1177/09544119251315877

Anal Chem. 2025 Feb 4. doi: 10.1021/acs.analchem.4c05271. Online ahead of print.

ABSTRACT

Lifetime determination plays a crucial role in fluorescence lifetime imaging microscopy (FLIM). We introduce UNET-FLIM, a deep learning architecture based on a one-dimensional U-net, specifically designed for lifetime determination. UNET-FLIM focuses on handling low photon count data with high background noise levels, which are commonly encountered in fast FLIM applications. The proposed network can be effectively trained using simulated decay curves, making it adaptable to various time-domain FLIM systems. Our evaluations of simulated data demonstrate that UNET-FLIM robustly estimates lifetimes and proportions, even when the signal photon count is extremely low and background noise levels are high. Remarkably, UNET-FLIM's insensitivity to noise and minimal photon count requirements facilitate fast FLIM imaging and real-time lifetime analysis. We demonstrate its potential by applying it to monitor rapid lysosomal pH variations in living cells during in situ acetic acid treatment, all without necessitating any modifications to existing FLIM systems.

PMID:39902564 | DOI:10.1021/acs.analchem.4c05271

Recent Pat Anticancer Drug Discov. 2025 Feb 3. doi: 10.2174/0115748928361472250123105507. Online ahead of print.

ABSTRACT

An aberrant increase in cancer incidences has demanded extreme attention globally despite advancements in diagnostic and management strategies. The high mortality rate is concerning, and tumour heterogeneity at the genetic, phenotypic, and pathological levels exacerbates the problem. In this context, lack of early diagnostic techniques and therapeutic resistance to drugs, sole awareness among the public, coupled with the unavailability of these modern technologies in developing and low-income countries, negatively impact cancer management. One of the prime necessities of the world today is the enhancement of early detection of cancers. Several independent studies have shown that screening individuals for cancer can improve patient survival but are bogged down by risk classification and major problems in patient selection. Artificial intelligence (AI) has significantly advanced the field of oncology, addressing various medical challenges, particularly in cancer management. Leveraging extensive medical datasets and innovative computational technologies, AI, especially through deep learning (DL), has found applications across multiple facets of oncology research. These applications range from early cancer detection, diagnosis, classification, and grading, molecular characterization of tumours, prediction of patient outcomes and treatment responses, personalized treatment, and novel anti-cancer drug discovery. Over the past decade, AI/ML has emerged as a valuable tool in cancer prognosis, risk assessment, and treatment selection for cancer patients. Several patents have been and are being filed and granted. Some of those inventions were explored and are being explored in clinical settings as well. In this review, we will discuss the current status, recent advancements, clinical trials, challenges, and opportunities associated with AI/ML applications in cancer detection and management. We are optimistic about the potential of AI/ML in improving outcomes for cancer and the need for further research and development in this field.

PMID:39902536 | DOI:10.2174/0115748928361472250123105507

J Bone Oncol. 2024 Dec 31;51:100659. doi: 10.1016/j.jbo.2024.100659. eCollection 2025 Apr.

ABSTRACT

BACKGROUND: Colorectal cancer is a prevalent malignancy with a significant risk of metastasis, including to bones, which severely impacts patient outcomes. Accurate prediction of bone metastasis risk is crucial for optimizing treatment strategies and improving prognosis.

PURPOSE: This study aims to develop a predictive model combining radiomics and Vision Transformer (ViT) deep learning techniques to assess the risk of bone metastasis in colorectal cancer patients using both plain and contrast-enhanced CT images.

MATERIALS AND METHODS: We conducted a retrospective analysis of 155 colorectal cancer patients, including 81 with bone metastasis and 74 without. Radiomic features were extracted from segmented tumors on both plain and contrast-enhanced CT images. LASSO regression was applied to select key features, which were then used to build traditional machine learning models, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest, LightGBM, and XGBoost. Additionally, a dual-modality ViT model was trained on the same CT images, with a late fusion strategy employed to combine outputs from the different modalities. Model performance was evaluated using AUC-ROC, accuracy, sensitivity, and specificity, and differences were statistically assessed using DeLong's test.

RESULTS: The ViT model demonstrated superior predictive performance, achieving an AUC of 0.918 on the test set, significantly outperforming all traditional radiomics-based models. The SVM model, while the best among traditional models, still underperformed compared to the ViT model. The ViT model's strength lies in its ability to capture complex spatial relationships and long-range dependencies within the imaging data, which are often missed by traditional models. DeLong's test confirmed the statistical significance of the ViT model's enhanced performance, highlighting its potential as a powerful tool for predicting bone metastasis risk in colorectal cancer patients.

CONCLUSION: The integration of radiomics with ViT-based deep learning offers a robust and accurate method for predicting bone metastasis risk in colorectal cancer patients. The ViT model's ability to analyze dual-modality CT imaging data provides greater precision in risk assessment, which can improve clinical decision-making and personalized treatment strategies. These findings underscore the promise of advanced deep learning models in enhancing the accuracy of metastasis prediction. Further validation in larger, multicenter studies is recommended to confirm the generalizability of these results.

PMID:39902382 | PMC:PMC11787686 | DOI:10.1016/j.jbo.2024.100659

Iran J Public Health. 2025 Jan;54(1):133-143. doi: 10.18502/ijph.v54i1.17583.

ABSTRACT

BACKGROUND: This study was conducted to perform deep learning-based body shape cluster analysis using 3D Body Scanner.

METHODS: For this study, 54 variables were measured using 3D Body Scanner on 366 adult men and women at Korea National Sport University in 2022. Transformer learning and dimensionality reduction models were used to perform cluster analysis on the measured data. Mann-Whitney test and Kruskal-Wallis test were applied to compare the principal component differences of new scale characteristics, and all statistical significance levels were set at .05.

RESULTS: First, among the two methods for classifying body types, the transformer algorithm had a higher performance in body type classification. Second, in the classification of body type clusters, two clusters, endomorphic body type and ectomorphic body type, were divided into six clusters, two for cluster 1 and four for cluster 2.

CONCLUSION: The six clusters provide more granular information than previous body type classifications, and we believe that they can be used as basic information for predicting health and disease.

PMID:39902372 | PMC:PMC11787839 | DOI:10.18502/ijph.v54i1.17583

Front Plant Sci. 2025 Jan 20;15:1514832. doi: 10.3389/fpls.2024.1514832. eCollection 2024.

ABSTRACT

Plant disease detection remains a significant challenge, necessitating innovative approaches to enhance detection efficiency and accuracy. This study proposes an improved YOLOv8 model, SerpensGate-YOLOv8, specifically designed for plant disease detection tasks. Key enhancements include the incorporation of Dynamic Snake Convolution (DySnakeConv) into the C2F module, which improves the detection of intricate features in complex structures, and the integration of the SPPELAN module, combining Spatial Pyramid Pooling (SPP) and Efficient Local Aggregation Network (ELAN) for superior feature extraction and fusion. Additionally, an innovative Super Token Attention (STA) mechanism was introduced to strengthen global feature modeling during the early stages of the network. The model leverages the PlantDoc dataset, a highly generalizable dataset containing 2,598 images across 13 plant species and 27 classes (17 diseases and 10 healthy categories). With these improvements, the model achieved a Precision of 0.719. Compared to the original YOLOv8, the mean Average Precision (mAP@0.5) improved by 3.3%, demonstrating significant performance gains. The results indicate that SerpensGate-YOLOv8 is a reliable and efficient solution for plant disease detection in real-world agricultural environments.

PMID:39902212 | PMC:PMC11788276 | DOI:10.3389/fpls.2024.1514832

Front Plant Sci. 2025 Jan 20;15:1502631. doi: 10.3389/fpls.2024.1502631. eCollection 2024.

ABSTRACT

Rice is an important part of the food supply, its different varieties in terms of quality, flavor, nutritional value, and other aspects of the differences, directly affect the subsequent yield and economic benefits. However, traditional rice identification methods are time-consuming, inefficient, and prone to damage. For this reason, this study proposes a deep learning-based method to classify and identify rice with different flavors in a fast and non-destructive way. In this experiment, 19 categories of japonica rice seeds were selected, and a total of 36735 images were finally obtained. The lightweight network High Precision FasterNet (HPFasterNet) proposed in this study combines the Ghost bottleneck and FasterNet_T0 and introduces group convolution to compare the model performance. The results show that HPFasterNet has the highest classification accuracy of 92%, which is 5.22% better than the original model FasterNet_T0, and the number of parameters and computation is significantly reduced compared to the original model, which is more suitable for resource-limited environments. Comparison with three classical models and three lightweight models shows that HPFasterNet exhibits a more comprehensive and integrated performance. Meanwhile, in this study, HPFasterNet was used to test rice with different flavors, and the accuracy reached 98.98%. The experimental results show that the network model proposed in this study can be used to provide auxiliary experiments for rice breeding and can also be applied to consumer and food industries.

PMID:39902203 | PMC:PMC11788896 | DOI:10.3389/fpls.2024.1502631

Biol Methods Protoc. 2024 Nov 20;10(1):bpae086. doi: 10.1093/biomethods/bpae086. eCollection 2025.

ABSTRACT

Brain age, as a correlate of an individual's chronological age obtained from structural and functional neuroimaging data, enables assessing developmental or neurodegenerative pathology relative to the overall population. Accurately inferring brain age from brain magnetic resonance imaging (MRI) data requires imaging methods sensitive to tissue health and sophisticated statistical models to identify the underlying age-related brain changes. Magnetic resonance elastography (MRE) is a specialized MRI technique which has emerged as a reliable, non-invasive method to measure the brain's mechanical properties, such as the viscoelastic shear stiffness and damping ratio. These mechanical properties have been shown to change across the life span, reflect neurodegenerative diseases, and are associated with individual differences in cognitive function. Here, we aim to develop a machine learning framework to accurately predict a healthy individual's chronological age from maps of brain mechanical properties. This framework can later be applied to understand neurostructural deviations from normal in individuals with neurodevelopmental or neurodegenerative conditions. Using 3D convolutional networks as deep learning models and more traditional statistical models, we relate chronological age as a function of multiple modalities of whole-brain measurements: stiffness, damping ratio, and volume. Evaluations on held-out subjects show that combining stiffness and volume in a multimodal approach achieves the most accurate predictions. Interpretation of the different models highlights important regions that are distinct between the modalities. The results demonstrate the complementary value of MRE measurements in brain age models, which, in future studies, could improve model sensitivity to brain integrity differences in individuals with neuropathology.

PMID:39902188 | PMC:PMC11790219 | DOI:10.1093/biomethods/bpae086