Background Neuro-ischemic diabetic foot ulcers (DFUs) associated with advanced peripheral artery disease and severe neuropathy pose a considerable challenge for limb preservation, especially when repeated revascularization efforts are deemed ineffective due to insufficient distal perfusion and a high risk of restenosis. This report discusses the case of an elderly male patient in his 60s, who was admitted for a non-healing wound following the amputation of the right great toe. Clinical examination revealed a 4.5 cm × 2.5 cm × 2 cm deep cavity exposing the first metatarsal head, with 50% slough coverage and substantial exudate. The right ankle-brachial index was measured at 0.52, and duplex ultrasound confirmed near-occlusive multisegmental disease of the anterior and posterior tibial arteries. Electromyography indicated severe peripheral neuropathy. The clinical diagnosis was determined to be Wagner grade 4, WIfI stage W2, I2, Fi grade 2, neuro-ischemic diabetic foot with arteriosclerosis obliterans. Objectives This study aims to assess the effectiveness of a limb-salvage protocol that integrates traditional Chinese incremental debridement, known as "silkworm-feeding," with negative-pressure wound therapy (NPWT) and autologous platelet-rich plasma (PRP) gel in a patient with a refractory neuro-ischemic diabetic foot ulcer (DFU), for whom further revascularization was not feasible. Methods After a multidisciplinary evaluation, the limb-salvage protocol was initiated. Incremental debridement was conducted to excise necrotic tissue while preserving viable, well-perfused tissue margins. A herbal ointment, referred to as "Shangke Huangyou Gao," was applied post-debridement to sustain a moist wound environment. Once healthy granulation tissue developed, NPWT and autologous PRP gel were sequentially integrated into the treatment plan. Results The total treatment course from the first hospitalization to complete healing extended over approximately four months, with full epithelialization occurring 9 weeks after discharge. At the 12-month follow-up, the foot remained intact. Conclusions In patients with refractory neuro-ischemic diabetic foot ulcers (DFU) for whom further revascularization is not feasible, the integrated approach of incremental debridement, negative-pressure wound therapy (NPWT), and platelet-rich plasma (PRP) may represent a feasible adjuncitve limb-salvage option in selected patients. Additional cases and prospective stuides are warranted to verify its safety, cost-effectiveness, and reproducibility.
Proper distinction between bacterial infection and sterile inflammation is one of the major issues in molecular imaging. Though radiolabelled leukocytes and [18F]fluorodeoxyglucose ([18F]FDG) have retained an important clinical role, neither is equally specific in a postoperative, inflammatory, and musculoskeletal environment. The lack of this requirement motivated the synthesis of antibiotic-based radiotracers, especially technetium-99m labelled fluoroquinolones. Initial interest in the directly labelled [99 mTc]Tc-ciprofloxacin was later challenged by radiochemical uncertainty, biodistribution variability and conflicting specificity data. The more recent activity has moved towards structurally defined derivative-based, as well as technetium(I)-tricarbonyl, nitrido, xanthate, and isonitrile platforms which can offer better radiochemical control and more reproducible physicochemical characterization. Ciprofloxacin- and norfloxacin-derived complexes have shown good radiochemical purity, in vitro stability and enhanced selectivity between bacterial infection and sterile inflammation in murine models. Nevertheless, these developments are not sufficient to achieve bacteria-specific targeting. Modern biological validation is too limited to establish a strong clinical role. The field has thus shown clear advancement on the aspect of radiopharmaceutical design, but the translational aspect is yet to finish. The key to future success will be characterized by strict structural characterization, mechanism-based biological testing, standardized comparative testing, and evidently improving clinically meaningful incremental value on the existing infection imaging tests.
This paper reviews recent advances in the precise management of prediabetes, adopting an integrative perspective that synergizes multi-omics, radiomics, and pharmacological insights from both traditional Chinese and Western medicine. In multi-omics research, genomics aids in early warning for high-risk populations among the Chinese; metagenomics reveals associations between intestinal microbiota characteristics and disease onset; metabolomics provides biomarkers, and the integration of multiple omics has identified novel prevention and treatment targets. Radiomics can assist in diagnosis and treatment, with integrated diagnostic models combining traditional Chinese and Western medicine expected to enhance diagnostic efficacy, while other imaging examinations can aid in early diagnosis, disease assessment, and treatment evaluation. In pharmacology, both traditional Chinese and Western drugs have their respective advantages and disadvantages, and their combined use can improve prevention and treatment outcomes. Integrated traditional Chinese and Western medicine treatment is theoretically supported, enabling the formulation of precise strategies, and clinical cases have confirmed its effectiveness. However, challenges remain in technological integration, standardization, and cost-effectiveness. In the future, with the development of technology, theory, artificial intelligence, and strengthened international cooperation, greater breakthroughs are anticipated.
Background Chemoradioresistance is one of the main causes of treatment failure and recurrence of gastric cancer (GC). Metabolic reprogramming, especially the abnormal activation of fatty acid oxidation (FAO), has been confirmed to be closely related to the chemoradioresistance of tumor cells. However, the role of very long-chain acyl-CoA dehydrogenase (VLCAD), the key rate-limiting enzyme of FAO, in the chemoradioresistance of GC and its upstream regulatory mechanism are still unclear. Methods The expression of VLCAD in GC cells was detected by western blot and qRT-PCR. Radiosensitivity, apoptosis and DNA damage were evaluated by colony formation assay, flow cytometry and immunofluorescence staining. Fatty acid metabolism was assessed by Oil Red O staining, FAO activity and ATP concentration detection. The effect of VLCAD on the radiosensitivity and metabolic regulation of GC in vivo was verified by xenograft tumor model. MeRIP assay, dual-luciferase reporter assay, RIP assay and actinomycin D assay were used to verify the modification and regulation of methyltransferase-like 3 (METTL3) and YTH domain family member 2 (YTHDF2) on VLCAD. Results VLCAD was downregulated in GC radioresistance cells (P < 0.05). After irradiation, VLCAD overexpression increased FAO activity and reduced lipid accumulation to promote radiosensitivity (P < 0.05). In vivo experiments further confirmed that overexpression of VLCAD could inhibit GC tumor growth and enhance the efficacy of radiotherapy (P < 0.05). In terms of mechanism, METTL3 inhibited VLCAD expression and mRNA stability in an m6A/YTHDF2-dependent manner (P < 0.05). In addition, METTL3 overexpression inhibited FAO activity and accelerated lipid accumulation to promote radioresistance and cisplatin resistance in GC cells by downregulating VLCAD (P < 0.05). Conclusion METTL3 reduces VLCAD expression by YTHDF2-dependent m6A modification, which decreases FAO activity, enhances lipid accumulation and ultimately promotes the chemoradioresistance of GC, providing a potential target for improving the chemoradiosensitivity of GC.
Background Liver fibrosis is a progressive disease that is progressive in nature and develops due to chronic liver injuries that can ultimately progress into cirrhosis and cancer of the liver if not detected early. Liver biopsies are a traditional method for diagnosing liver diseases, but this approach is invasive and expensive, among other drawbacks. Objective An approach called LivoScan, which is a hierarchical attention-guided deep learning framework that automatically identifies and classifies liver fibrosis using MRI and ultrasound images, is proposed in this research. Methods This method incorporates convolutional-based feature extraction with an attention-guided learning representation approach in order to achieve better fibrosis-related features discrimination. The effectiveness of the proposed approach was tested on two datasets, the CHAOS MRI database and the Liver Histopathology Fibrosis Ultrasound Images Database (LHFUI). Several methods, including accuracy, precision, recall, F1-score, Area Under the Curve (AUC), kappa coefficient, confusion matrix, Receiver Operating Characteristic (ROC) analysis, etc., were considered. Results Despite the impressive results, there is a possibility that the network learned specific features from the datasets that may not generalize well to other cases. The lack of sufficient data diversity and class imbalance are some issues to consider. However, the proposed network architecture showed significantly better results compared to several other methods based on deep learning. Conclusion The proposed LivoScan system demonstrated excellent classification results on both the CHAOS MRI and LHFUI ultrasound datasets, showing its suitability for evaluating liver fibrosis non-invasively by means of various imaging techniques. More research with bigger datasets needs to be conducted.
Radiation therapy is a key treatment modality for cancer, effectively targeting malignant cells. However, the treatment is often associated with harmful side effects due to radiation-induced damage to healthy tissues, posing a significant clinical challenge. Curcumin, a naturally occurring polyphenolic compound in turmeric, has attracted considerable interest for its radioprotective potential. This review examines the diverse roles of curcumin in mitigating radiation-induced damage in cancer patients, with a focus on its molecular and biochemical mechanisms. Curcumin's antioxidant properties help scavenge reactive oxygen species (ROS), reducing oxidative stress, while its anti-inflammatory effects modulate cytokine responses, further protecting normal tissues. Additionally, curcumin enhances DNA repair mechanisms, ensuring genomic stability, and regulates apoptotic pathways to prevent excessive cell death. Preclinical and in vitro studies have provided encouraging evidence of curcumin’s efficacy in reducing radiation-induced toxicity without compromising therapeutic outcomes. Ongoing research seeks to refine curcumin’s clinical application, with emphasis on optimizing dosage, improving bioavailability, and exploring combination approaches with other therapies. Despite these promising findings, challenges remain, including the need for reliable biomarkers to predict patient response and the development of more efficient curcumin delivery systems. Although substantial evidence supports the radioprotective effects of curcumin in cellular and animal models, clinical evidence remains limited and is currently restricted to a small number of studies evaluating radiation-induced toxicities, including oral mucositis, dermatitis, and pneumonitis. Therefore, the translational potential of curcumin should be interpreted cautiously until larger randomized controlled trials establish optimal dosing regimens, formulations, long-term safety, and effects on tumour control outcomes. In conclusion, Curcumin represents a promising adjunctive radiomodulator; however, current evidence remains predominantly preclinical, and further well-designed clinical trials are necessary before routine integration into radiation oncology practice. Further research is required to address current limitations and advance the translation of curcumin into clinical practice. Regulatory considerations and patient education will be essential in integrating curcumin supplementation into standard cancer care protocols to improve patient outcomes.
Background PSMA-targeted radiotheranostics has become an important component of precision prostate cancer management, supported by recent clinical trials of ^177Lu-PSMA-617 and ^225Ac-PSMA ligands. Concurrently, the therapeutic landscape of metastatic prostate cancer has expanded to include androgen receptor pathway inhibitors, chemotherapy, and PARP inhibitors. Despite these advances, treatment response remains heterogeneous, and current radiotheranostic workflows remain fragmented across imaging, radiopharmaceutical design, dosimetry, and clinical implementation. Objectives This study proposes a convergence-based methodological framework that integrates molecular imaging, biomarker stratification, nanotechnology-enabled delivery, and artificial intelligence (AI)-driven dosimetry into a unified and adaptive radiotheranostic strategy for prostate cancer. Methods A focused synthesis of recent clinical and translational studies (2015–2025) was performed covering PSMA PET phenotyping, radiopharmaceutical engineering, nanoplatform-based delivery systems, biomarker-guided isotope selection, and AI-enabled dosimetry. These components were integrated into a multi-domain convergence model linking ^68Ga-PSMA PET imaging, molecular biomarkers, nanotechnology platforms, and predictive computational analytics into a continuous decision-support pipeline. Results The proposed framework demonstrates how quantitative PSMA PET metrics and molecular biomarkers can guide rational selection between β- and α-emitting radionuclides and enable adaptive dose optimization. Integration of nanoplatforms improves tumor targeting and therapeutic flexibility, while AI-driven predictive models support estimation of tumor response and organ toxicity prior to radionuclide administration. Evidence from recent preclinical and clinical studies supports the feasibility of image-guided personalization and adaptive treatment strategies within this convergence architecture. Conclusions The convergence-based framework advances PSMA radiotheranostics from isolated technological components toward a biologically informed, data-driven, and clinically scalable therapeutic paradigm. By unifying molecular imaging, biomarkers, nanotechnology, and AI-driven dosimetry, this approach addresses limitations of population-based dosing and heterogeneous treatment response, providing a foundation for next-generation precision radiotheranostic medicine.
Cardiovascular disease is the leading cause of morbidity and mortality in patients with type 2 diabetes mellitus (T2DM). Early detection of subclinical coronary and myocardial injury is therefore essential to reduce future cardiovascular events. MPI SPECT and its gated form (GSPECT MPI) provide simultaneous assessment of three-dimensional tracer distribution and functional/mechanical cardiac parameters, enabling detection of ischemia, viable myocardium, left-ventricular remodeling, systolic and diastolic dysfunction, and mechanical dyssynchrony often before symptoms arise. Across diverse cohorts, abnormal perfusion on SPECT consistently identifies T2DM patients at substantially higher risk of major adverse cardiovascular events (MACE), whereas normal perfusion generally corresponds to low short- to mid-term event rates. GSPECT-derived metrics including post-stress LVEF and change in LVEF, EDV/ESV, transient ischemic dilation (TID), lung-to-heart ratio, peak filling rate (PFR), and phase-dispersion measures frequently provide independent, incremental prognostic information and can refine risk assessment even when perfusion is preserved. Clinically actionable patterns (abnormal MPI, marked post-stress LVEF decline, elevated TID or lung-to-heart ratio, reduced PFR, and pronounced phase dispersion) should prompt intensified risk-factor modification, closer surveillance, and, when appropriate, anatomic coronary evaluation. Methodological heterogeneity in stress protocols, gating parameters, reconstruction algorithms, and analysis software limits the adoption of universal thresholds and underscores the need for methodological standardization and local validation. This narrative review aims to synthesize current evidence on SPECT MPI and GSPECT applications in T2DM to inform risk-tailored, targeted use of these modalities for early detection and personalized management of subclinical cardiovascular disease.
Introduction/objective: Patients undergoing diagnostic nuclear medicine procedures temporarily become sources of external radiation, potentially contributing to exposure of staff, accompanying persons, and other patients. This study aimed to evaluate activity-normalized survey-meter readings around patients undergoing routine 99(m)Tc-pertechnetate thyroid scintigraphy as a function of distance, anatomical level, and post-injection measurement time under real clinical conditions. Methods: A prospective observational study was conducted in eight patients. Free-in-air survey-meter readings were obtained using a NEB.211 dose-rate meter equipped with an energy-compensated Geiger-M & uuml;ller detector. Measurements were performed at four anatomical levels - head, thorax, abdomen, and pelvis - and five distances - 25, 50, 100, 150, and 200 cm - on three post-injection measurement occasions. Because measurement timing varied under routine workflow conditions, analyses were based on recorded actual times, with mean post-injection times of 19.6 +/- 14.4 min, 54.4 +/- 17.1 min, and 91.5 +/- 14.8 min. Readings were normalized to administered activity and interpreted as comparative survey-meter readings rather than calibrated H*(10) or Hp(10) dose-equivalent quantities. The distance-reading relationship was evaluated using log-log regression. Results: Activity-normalized survey-meter readings decreased consistently with increasing distance at all anatomical levels and across all three measurement occasions. Log-log regression demonstrated a strong inverse relationship between distance and activity-normalized readings, with empirical distance exponents ranging from-0.746 to-0.975 and corresponding R2 values ranging from 0.993-0.999. Abdominal and thoracic levels generally showed higher readings than head and pelvic levels, whereas temporal changes within the observed interval were modest compared with the distance effect. Discussion: The findings indicate that the patient-related radiation field under routine clinical conditions is strongly distance dependent and influenced by anatomical measurement level. Because the instrument did not have a traceable H*(10) calibration certificate, the results should be interpreted as comparative activity-normalized survey-meter data rather than absolute ambient dose-equivalent measurements. Conclusion: Increasing distance was the dominant practical factor for reducing measured patient-related radiation levels around patients undergoing 99mTc-pertechnetate thyroid scintigraphy. These findings support simple distance-based workflow optimization measures in nuclear medicine departments.
Background: Alpha-emitting radionuclides, characterized by high-linear energy transfer (LET) and short path length, are emerging as potent tools in targeted cancer therapy. When combined with advanced nanocarrier platforms, they enable precise tumor targeting, improved pharmacokinetics, and integrated diagnostic therapeutic (theranostic) applications. Objective: This review aims to comprehensively evaluate the current status and future prospects of integrating alpha emitters with nanocarrier systems for precision oncology, with a focus on technological innovations, clinical translation, and regulatory considerations. Methods: A targeted literature search (2024-2025) was conducted across PubMed, Scopus, and Web of Science, emphasizing peer-reviewed studies from high-impact journals. Key thematic areas include alpha-emitter characteristics, nanocarrier design, radiochemistry, tumor-targeting ligands, preclinical/clinical data, and manufacturing challenges. Results: Commonly investigated alpha-emitters such as "225Ac, "211At, "213Bi, and "212Pb exhibit superior cytotoxic profiles compared to beta-emitters, particularly in micro metastatic and resistant cancers. Nanocarrier systems liposomes, dendrimers, polymeric nanoparticles, and exosomes offer tunable physicochemical properties, ligand functionalization, and enhanced in vivo stability. Advances in chelation chemistry and cleavable linker strategies improve radiolabelling stability and isotopic purity. Preclinical studies demonstrate enhanced tumor uptake and reduced systemic toxicity, while early-phase clinical trials highlight safety and therapeutic promise. However, GMP-scale isotope production, dosimetry standardization, and regulatory harmonization remain critical bottlenecks. Conclusion: Alpha-emitter Nanotheranostics represent a transformative approach in nuclear medicine, combining high-LET radiotoxicity with precision delivery. Overcoming manufacturing, regulatory, and logistical challenges will be essential for widespread clinical adoption. Emerging
Background: Theranostics is an innovative concept in nuclear medicine that combines molecular imaging with targeted radionuclide therapy. PSMA is the best-studied and most clinically utilized oncological theranostic target, as it is a protein that is highly and differentially expressed in latestage prostate cancer. Objective: This review comprises a comprehensive appraisal of PSMA-targeted radiopharmaceuticals and emphasizes their physiological basis, molecular blueprint, clinical imaging, and treatment utilization. Methods: A narrative review of peer-reviewed literature was conducted, encompassing preclinical studies, prospective clinical trials, meta-analyses, and real-world evidence on PSMAtargeted imaging and radionuclide therapy. Emphasis was placed on clinically validated positron emission tomography tracers and therapeutic agents, including beta- and alpha-emitting radiopharmaceuticals. Results: PSMA-targeted positron emission tomography imaging with gallium-68 and fluorine-18labelled tracers provides excellent diagnostic performance in both staging and restaging scenarios and is therefore pivotal for patient selection in radioligand therapy. PSMA radioligand therapy with 1 77Lu-PSMA-617 has demonstrated substantial benefits in progression-free and overall survival in metastatic castration-resistant prostate cancer, while maintaining a good safety profile. Nevertheless, treatment response is variable, mainly due to differences in PSMA expression, disease biology, and dosimetric factors. Recent data from the past few years also indicate that personalized dosimetry, the use of alpha-emitting compounds, and combinations of treatment regimens, such as Poly(ADP-ribose) polymerase (PARP) inhibitor and immunotherapy, could further improve clinical outcomes; however, challenges such as long-term toxicity and shortages of radionuclides must be addressed. Conclusion: PSMA-targeted theranostics are a clinically proven platform for precision oncology. Subsequent developments will rely on biologically guided patient stratification, treatment personalization, refined dosimetry, careful clinical evaluation of novel radiopharmaceuticals, and long-term safety assessment.
Background: Radioguided surgery (RGS) using PSMA radioligands emerged as a valuable approach for the detection of metastatic lymph nodes in prostate cancer. [Tc-99(m)]Tc-PSMA-I&S is specifically designed for combined imaging and RGS, however, its clinical implementation as an experimental radiopharmaceutical remains limited by heterogeneous radiolabeling procedures, particularly in automated preparation settings. Objectives: To optimize the preparation conditions of [Tc-99(m)]Tc-PSMA-I&S and develop an automated synthesis protocol. Methods: A systematic study of the PSMA-I&S radiolabeling conditions with Tc-99(m) was conducted using manual preparations. Key parameters including buffer type and concentration, transfer ligand, antioxidant compound, precursor amount, and heating time were evaluated based on radiochemical purity (RCP) assessed by radio-HPLC. Optimized conditions were subsequently transposed to a fully-automated synthesis on a GAIA (R) module, following a workflow based on Ga-68 radiopharmaceuticals. Three test batches were produced and subjected to extensive quality controls, including RCP, radiochemical yield (RCY) and stability. Results: Manual optimization identified sodium formate buffer (1.5 M, pH 8.5), stannous chloride as the sole reducing agent, and 95 degrees C heating for 5 min as optimal conditions, yielding RCP values >= 98% without requiring any transfer ligand or antioxidant. These conditions were successfully implemented on the synthesizer. Automated production of [Tc-99(m)]Tc-PSMA-I&S (n = 3) was achieved within 24.12 +/- 0.20 min, starting from 3.31 +/- 0.18 GBq of Tc-99(m), with RCY of 72.3 +/- 6.7% and RCP of 95.27 +/- 0.21%. All batches remained stable over 6 h, maintaining RCP > 95%. Conclusion: This work establishes a fast, simplified and fully automated radiolabeling process for [Tc-99(m)]Tc-PSMA-I&S. The proposed methodology supports GMP-compliant production and would facilitate the broader implementation of PSMA-targeted RGS.
Background: Radiation-induced lung injury (RILI) is damage to normal lung tissue caused by thoracic tumor radiotherapy (RT), but effective interventions are lacking and its molecular mechanisms remain unclear. Objectives: This study aimed to investigate the role and regulatory mechanism of endothelial cell-specific molecule 1 (ESM1) in RILI and to explore the involvement of the GATA binding protein 3 (GATA3)/ESM1/nuclear factor kappa B (NF-kappa B) signaling axis in disease progression. Methods: The gene expression omnibus (GEO), search tool for the retrieval of interacting genes (STRING), JASPAR, and ensembl databases were used to obtain RILI-related information. Cell transfection was performed for gene knockdown and overexpression. Western blot, quantitative polymerase chain reaction (qPCR), and enzyme-linked immunosorbent assay (ELISA) were used to detect protein, messenger RNA (mRNA), and inflammatory factor expression levels, respectively. Dual-luciferase reporter assay and chromatin immunoprecipitation (ChIP) were conducted to verify the binding between transcription factors and promoters. Hematoxylin-eosin (HE) staining and masson's trichrome staining (Masson) were used to evaluate tissue lesions and pulmonary fibrosis. Results: ESM1 was upregulated in RILI. Under radiation, knockdown of ESM1 inhibited NF-kappa B signaling and activated inhibitor of NF-kappa B alpha (I kappa B alpha) expression, while reduced the levels and expression of tumor necrosis factor-alpha (TNF-alpha),interleukin-6 (IL-6),IL-8,transforming growth factor-beta1 (TGF-beta 1),diacylglycerol (DAG),protein kinase C theta (PKC theta) and B-cell CLL 10 (Bcl 10), which was confirmed in RILI mouse models. The transcription factor GATA binding protein 3 (GATA3) bound the ESM1 promoter and activated its expression; knockdown of GATA3 showed effects similar to ESM1 knockdown, while overexpression of GATA3 restored the effects of ESM1 knockdown. HE staining showed that knockdown of GATA3 alleviated lung tissue lesions in RILI, Masson showed reduced pulmonary fibrosis, and knockdown of GATA3 also decreased respiratory rate and pulmonary edema and mitigated body weight loss in RILI mouse models. Conclusions: The transcription factor GATA3 positively regulate ESM1, thereby modulating the NF-kappa B pathway and the inflammatory response, which in turn affect the progression of RILI.
Objective: This study aims to address the challenges inherent in thyroid nodule ultrasound image segmentation, such as excessive computational complexity, limited data availability, and the suboptimal segmentation performance of traditional deep learning models. To this end, we propose a lightweight U-shaped segmentation network named SCDA-Net, specifically designed to balance segmentation efficiency and accuracy while minimizing computational costs. By optimizing the network structure and parameters, SCDA-Net achieves more precise segmentation of thyroid nodules, thereby improving the accuracy and efficiency of clinical diagnosis. Methods: This study utilized the TN3K dataset, which comprises 3493 ultrasound images. First, all ultrasound images were cropped and scaled to a standardized size of 224 & times; 224 pixels. Then, the dataset was randomly partitioned into training, testing, and validation sets in a ratio of 7:1:2. Furthermore, to mitigate the issue of insufficient training samples, data augmentation techniques including translation, random flipping, and Gaussian noise addition were employed. Subsequently, the preprocessed training set was used to train the lightweight U-shaped SCDA-Net. Finally, the performance of the trained model was evaluated on the test set. Conclusion: Through the aforementioned technical advances and innovations, this study provides a robust and efficient solution for the computer-aided diagnosis of thyroid nodules using ultrasound images, which significantly improves diagnostic accuracy. These findings hold far-reaching significance in clinical practice, as it not only facilitates the optimization of patient treatment plans but also contribute to the overall advancement of medical technology.
Background and purpose: Accurate recognition of emotional states is critical for monitoring neuropsychiatric disorders, however, deploying such systems on battery-powered wearable devices necessitates exceptional energy efficiency. This study aims to develop a lightweight, bioinspired framework for energy-efficient multimodal emotion recognition. Methods: We propose EAVS-TF, a Spiking Neural Network (SNN) framework integrating three key components: (1) a biologically motivated adaptive feature selection module that utilizes short-term leaky integration and sparse gating to dynamically recalibrate audio-visual channels; (2) a compact Transformer encoder for deep cross-modal fusion; and (3) an energy-efficient Leaky Integrate-and-Fire (LIF)-based spiking classifier. The model was evaluated using the CREMA-D benchmark dataset. Results: EAVS-TF achieved a recognition accuracy of 81.48%, outperforming previous SNN-based methods by 3-11% points. Ablation studies confirmed that the adaptive feature selection mechanism significantly enhanced both robustness and fusion quality. Conclusions: EAVS-TF successfully synthesizes the event-driven efficiency of SNNs with the representational power of Transformers. The framework demonstrates significant potential as a low-power, non-invasive tool for continuous behavioral monitoring in future wearable and neuromorphic systems
Thoracic and abdominal tumors, such as lung, breast, liver, and colorectal cancers, are characterized by high incidence and mortality rates, posing a significant threat to global health. Traditional diagnostic and therapeutic approaches, primarily based on histopathological assessment and single-modality imaging, often fail to capture tumor heterogeneity fully. In the era of precision medicine, the integration of multi-omics technologies (genomics, transcriptomics, proteomics, metabolomics) with multimodal imaging (computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET)) has emerged as a promising strategy to bridge macroscopic imaging phenotypes and microscopic molecular characteristics. This review systematically explores the synergistic application of multi-omics analysis and multimodal imaging in diagnosing and treating thoracic and abdominal tumors. It highlights advances in imaging technologies (such as radiomics, artificial intelligence (AI)-assisted interpretation) and multi-omics approaches (such as liquid biopsy and single-cell sequencing), alongside their clinical applications in early diagnosis, molecular subtyping, treatment response evaluation, and prognosis prediction across thoracic-abdominal and other specialties such as liver cancer and cardiothoracic surgery. Despite challenges in data standardization, algorithm interpretability, and clinical translation, this integrated framework holds great potential for advancing personalized oncology. Future directions should focus on leveraging next-generation technologies (such as photon-counting CT and spatial transcriptomics), transitioning from correlative analyses to causal mechanistic studies, and translating integrated imaging-omics models into clinical decision support systems to guide personalized therapy.
Accurate calculation of dose distribution is a critical component in intensity-modulated radiation therapy (IMRT). The dose distribution exhibits significant variability across patients. To address this challenge, we propose a deep learning-based dose prediction method specifically designed for brain tumors in the context of IMRT. By incorporating the anatomical characteristics of brain tumors, we developed an innovative convolutional neural network architecture. Furthermore, we employed a structural similarity loss function to enhance the precision of dose distribution predictions for brain tumor patients. Experimental results demonstrate its superior performance over state-of-the-art methods in terms of Mean Absolute Error (MAE) and Dose Volume Histogram (DVH) metrics.
Objective Building on accumulating evidence that atopic inflammation and neuroimmune/neurocognitive processes may influence the central nervous system, we aimed to compare structural brain imaging features between asthma patients with and without atopic dermatitis (AD), and to examine their associations with asthma symptom-related clinical measures using structural MRI. Methods Thirty asthma patients with AD (Group A), 30 asthma patients without AD (Group B), and 30 healthy controls (Group C) were recruited. All participants underwent sMRI scanning. Voxel-based morphometry (VBM) and surface-based morphometry (SBM) analyses were performed to examine structural differences. Brain regions showing significant differences were further analyzed for correlations with AD- and asthma-related clinical indicators. Results Baseline demographic characteristics did not differ significantly among the three groups. Asthma-related clinical indices showed no significant differences between Groups A and B. VBM analysis revealed no significant differences in total intracranial volume (TIV), gray matter volume (GMV), white matter volume (WMV), or cerebrospinal fluid (CSF) volume. Similarly, SBM analysis detected no significant differences in cortical surface area (CSA), cortical sulcal curvature (CSC), or sulcal depth (SD). However, significant cortical thickness (CT) reductions were identified in several regions, including the left middle frontal gyrus, left supplementary motor area (SMA), left orbitofrontal cortex (OFC), right primary visual cortex, right secondary visual cortex, and right OFC. No significant correlations were found between VBM/SBM parameters and the AD index in Group A, or asthma indices in Groups A and B Conclusion Asthma patients exhibited structural brain alterations, with more pronounced cortical thinning in those with comorbid AD compared to asthma-only patients. These findings suggest that cortical thinning may reflect underlying neurocognitive mechanisms in asthma and could serve as a potential neuroimaging biomarker for central nervous system involvement in atopic disease.
Background: Several studies have reported that benzimidazole-based structures are used in the diagnosis of Alzheimer's disease. Various modified benzimidazole-based structures serve as radiotracers. Among them, 2-(4-(2-(fluoro-18F)ethyl)piperidin-1-yl)benzo[4,5]imidazo[1,2-a]pyrimidine or [18F]T808 is a prominent radiotracer for tau PET tracer. The synthesis was started with trichloroacetyl chloride and ethyl vinyl ether, the precursor synthesis was completed in 9 steps. The precursor facilitated the radiosynthesis of [18F]T808 in excellent yield with higher purity. Objective: We explored a radiotracer for diagnosing Alzheimer's disease, specifically a benzimidazole-based tau PET precursor. The goal was to synthesise and develop a semi-automated radiolabelling protocol of 2-(4-(2-(fluoro-18F)ethyl)piperidin-1-yl)benzo[4,5] imidazo[1,2-a] pyrimidine, resulting in a higher yield with shorter radiosynthesis time. Methods: The synthesis of the precursor started with Ethyl vinyl ether and trichloroacetyl chloride. All chemically synthesised compounds were characterised by 1H-NMR, 13C-NMR, and mass spectrometry. The synthesised precursor was used for the radiosynthesis of tau PET tracer [18F]T808 by the nucleophilic [18F]fluorination with K[18F]F/Kryptofix 2.2.2 in DMSO at 140 degrees C. This transformative process resulted in a crude radiolabeled product being purified through semipreparative high-performance liquid chromatography (HPLC) and solid phase extraction (SPE) in an isolated desired product. The synthesised radiotracer was analysed using analytical tools such as radio TLC, HPLC, pH, endotoxin, and half-life. Results: The precursor was successfully synthesised in 9 crucial steps with 80% yields, and '97% chemical purity. The synthesised precursor was used in the semi-automated radiosynthesis of [18F]T808. The radiolabelled desired product [18F]T808 was successfully achieved. The decaycorrected yield for [18F]T808 was approximately 45-50% at the end of synthesis, 40-45 min Conclusion: Our method resulted in tau PET tracer precursor synthesis of 65% yield with 97% chemical purity. The tau tracer was radiolabeled with 45-50% radiochemical purity of '98%. The analytical data match regulatory standards for being used as a PET tracer for clinical studies.
Objective Accurate preoperative prediction of central lymph node metastasis (CLNM) in papillary thyroid carcinoma (PTC) remains challenging. This study sought to develop and validate a combined model that combines ultrasound radiomics and proteomic features to predict CLNM in PTC patients as accurately as possible. Methods A retrospective study was conducted on 155 PTC patients. Two experienced radiologists manually delineated the regions of interest (ROIs) on preoperative ultrasound images using ITK-SNAP, and 1322 radiomic features were extracted from each ROI using the PyRadiomics package, encompassing first-order, texture, and wavelet-based features. Proteomic profiling was then performed on formalin-fixed paraffin-embedded tissue samples via data-independent acquisition mass spectrometry. Feature selection was additionally performed using Pearson correlation and LASSO regression. Machine learning models were built using support vector machine, logistic regression, random forest, and XGBoost based on radiomics, proteomics, and combined features. Results The combined model significantly outperformed single-modality models, achieving an area under the curve of 0.84 in the validation set, compared to 0.75 (radiomics) and 0.75 (proteomics). Decision curve analysis confirmed superior clinical utility as well. Functional enrichment analysis revealed upregulation of metabolic pathways and downregulation of immune-related extracellular matrix organization in CLNM-positive patients. Conclusion Integrating ultrasound radiomics and proteomics provides a powerful framework for predicting CLNM in PTC. Although its results are based on surgical specimens, this study lays a methodological foundation and provides a theoretical framework for future proteomic biomarker research using preoperative fine-needle aspiration biopsies, thereby facilitating personalized treatment strategies and offering insights into the underlying biology of metastasis.
