BACKGROUND:Major depressive disorder (MDD) ranks among the foremost contributors to disability worldwide, yet its neurophysiological mechanisms remain poorly understood. Excitation-inhibition (E/I) imbalance has been implicated in MDD pathophysiology, but cortex-wide E/I ratio and its molecular substrates in MDD remain unknown. METHODS:Resting-state functional magnetic resonance imaging data from 254 MDD patients and 451 healthy controls (HCs) across six sites were analyzed. The Hurst exponent, a biophysically confirmed proxy of E/I balance, was estimated using a fractionally integrated process framework. Neurobiological decoding analyses were performed to map the transcriptomic and neurochemical signatures of cortical E/I imbalance in MDD. An independent ketamine clinical trial dataset (32 treatment-resistant depression patients and 21 HCs) was used to examine ketamine-induced changes in cortical E/I balance. RESULTS:Patients with MDD demonstrated significantly reduced Hurst exponent values, predominantly encompassing the parietal and prefrontal-cingulate cortices. Transcriptomic analysis identified enrichment for neuronal structural organization, nucleic acid metabolism, and mitochondrial function, with preferential overlap with excitatory and inhibitory neuron-specific gene sets. Neurochemically, Hurst exponent alterations were spatially associated with GABAergic, opioidergic, serotonergic, and synaptic density distributions. Divergent group-by-treatment effects were observed in the anterior cingulate and medial prefrontal cortices, with ketamine-induced increases in TRD patients. CONCLUSION:These findings highlight that prefrontal-cingulate E/I imbalance, anchored to specific transcriptional and neurochemical substrates, may underlie the pathophysiology of MDD and the antidepressant effects of ketamine. The Hurst exponent offers a promising neuroimaging approach for probing E/I imbalance and identifying potential treatment targets in depression.

Migraine is a prevalent and debilitating neurological disorder with poorly understood neural mechanisms. Here, we characterize a hypothalamic-trigeminal pathway involved in the regulation of migraine-like allodynia. Using a combination of monosynaptic circuit, fiber photometry-based calcium imaging, and behavioral assays in a nitroglycerin (NTG)-induced murine model, we observed hyperactivity in corticotropin-releasing factor (CRF)-expressing neurons within the hypothalamic paraventricular nucleus (PVN). Activation or inhibition of PVNCRF neurons mimicked or blocked migraine-like allodynia, respectively. These PVNCRF neurons modulated migraine-like allodynia by exciting glutamatergic neurons in the spinal trigeminal nucleus caudalis (SP5C). Furthermore, employing a CRF neurotransmitter fluorescent sensor, neuropharmacology, and electrophysiological recordings, we revealed that PVNCRF neurons excessively release CRF neuropeptides onto SP5CGlu neurons during migraine-like conditions. This led to hyperactivation of corticotropin-releasing factor receptor type 2 (CRFR2), but not type 1 (CRFR1), resulting in hyperalgesia. Blockade of CRFR2 within the SP5C significantly alleviated migraine-like allodynia. Complementing these findings, clinical functional magnetic resonance imaging (fMRI) of migraine patients indicated structural and functional alterations in PVN and SP5C regions associated with this pathway. Collectively, our results uncover a previously unappreciated PVNCRF-SP5CGlu pathway in migraine-like allodynia, providing novel insights into the neurobiology of migraine and identifying potential therapeutic targets.

Purpose To examine whether time-dependent diffusion MRI (Td-dMRI) and macromolecular proton fraction (MPF) mapping-derived quantitative metrics can effectively distinguish between cervical cancer with and without lymph node metastasis (LNM) before treatment. Materials and Methods In this prospective study of adults with clinically suspected cervical cancer who underwent Td-dMRI, MPF mapping, and pulsed gradient spin-echo diffusion-weighted imaging (DWIPGSE) examinations between October 2023 and June 2025, authors calculated Td-dMRI-derived parameters (cellularity, diameter, intracellular volume fraction [Vin], and extracellular diffusivity [Dex]), MPF, and DWIPGSE-derived parameter (pulsed gradient spin-echo apparent diffusion coefficient [ADCPGSE]). Through Ridge regression analysis, the authors identified independent predictors of LNM and developed a composite diagnostic tool using logistic regression analysis. To evaluate tool performance, the area under the receiver operating characteristic curve was determined. Results Among 98 female individuals with cervical cancer (mean age, 56.69 years ± 11.63 [SD]), participants who were LNM positive exhibited higher cellularity, Vin, and MPF but lower diameter, Dex, and ADCPGSE than their counterparts who were LNM negative (P < .001 to P = .007). Cellularity, maximum tumor diameter, and MPF were independent predictors of LNM status, with their combination yielding the best diagnostic performance (area under the receiver operating characteristic curve, 0.95; 95% CI: 0.89, 0.98). The performance of this combination surpassed that of individual imaging modality, including DWIPGSE (ADCPGSE), and MPF, as well as any individual parameter, including cellularity, Vin, diameter, and Dex. Conclusion Td-dMRI and MPF mapping were effective for predicting LNM in cervical cancer, with the combination of cellularity, maximum tumor diameter, and MPF showing the best diagnostic performance. Keywords: Time-Dependent Diffusion MRI, Macromolecular Proton Fraction, Cervical Cancer, Lymph Node Metastases © RSNA, 2026.

Purpose:Hafnium oxide nanoparticles have been established as effective radiosensitizers, however, tumor cells often develop resistance to single-modality radiotherapy, and the tumor microenvironment (TME) poses additional limitations to treatment efficacy. To address these challenges, we fabricated a doxorubicin and manganese oxide co-loaded hafnium oxide (MD-Hf) nanoplatform for synergistic radio-chemotherapy and evaluated its antitumor performance in cellular and animal models. Results:In MD-Hf nanoplatform, the HfO2 nanocrystal functions as the radiosensitizer and carrier, the Dox works for chemotherapy, while the manganese oxide coating layer are capable of modulating the TME by depleting glutathione (GSH) and converting H2O2 in to ·OH radicals. Moreover, the MnOx coating also allows the nanoplatform possessing TME-responsive Dox release. Upon exposure to X-rays, the MD-Hf exhibited evident toxicity to Panc 02 tumor cells, only 63.9±10.7% cell remain alive after irradiated with 2Gy X-ray, which is much lower than 86.7±6.33% of the group administrated with pure HfO2 NPs. In vivo studies further demonstrated superior therapeutic outcomes with the MD-Hf nanoplatform, as evidenced by markedly reduced tumor size and weight compared to treatment with HfO2 nanoparticles alone. RNA-seq analysis reveals the Dox can potentiate organelle damage, and the MnOx can even activate immune response, which further corroborates the multifunctionality of the integrated nanoplatform. Conclusion:The newly developed doxorubicin and manganese oxide co-loaded HfO2 nanoplatform significantly enhance radio-chemotherapeutic efficacy against pancreatic tumor cells, offering a promising strategy that may guide the future clinical development of HfO2-based radiotherapy.

BACKGROUND:Major depressive disorder (MDD) has been characterized by widespread functional network alterations. However, previous neuroimaging studies have predominantly focused on inter-regional functional connectivity, overlooking intrinsic temporal hierarchies that reflect fundamental neural processing properties. METHODS:We quantified intrinsic neural timescales (INT), a measure of temporal autocorrelation in intrinsic neural activity, across cortical and subcortical structures using resting-state functional MRI in patients with MDD and controls. We also assessed the associations between the INT alterations and depressive symptom and age effects. To elucidate neurobiological substrates of the INT alterations, we examined spatial associations with transcriptomic and neurotransmitters profiles. RESULTS:Patients with MDD exhibited widespread INT reductions across bilateral cingulate, insular, and frontoparietal cortices and subcortical structures. The association between age and INT differed between groups, with patients showing age-related INT increases in the posterior cingulate cortex. Transcriptomic analyses identified gene signatures enriched for synaptic transmission, glutamatergic pathways, and ion channel regulation, showing significant overlap with astrocyte- and excitatory-neuron-specific markers. Neurotransmitter mapping revealed associations spanning serotonergic, GABAergic, glutamatergic, and cholinergic receptor systems. CONCLUSION:These findings reveal widespread INT reductions in MDD and their spatial correspondence with normative transcriptomic and neurotransmitter profiles. The co-localization of INT alterations with normative gene expression enriched for synaptic and glutamatergic pathways provides insights into the molecular context of intrinsic temporal processing abnormalities in MDD.

To develop and compare multiregional 2D and 2.5D deep learning models based on DCE-MRI for noninvasive prediction of axillary lymph node (ALN) pathological complete response (pCR) after neoadjuvant chemotherapy (NAC). This retrospective study enrolled 305 patients with invasive breast cancer and ipsilateral ALN metastasis, which were randomly assigned to a training set (n = 214) and a validation set (n = 91). The Mann–Whitney U test, Spearman correlation analysis, max-relevance and min-redundancy and least absolute shrinkage and selection operator were used to select the most significant features. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curve analysis, and decision curve analysis. Among the 305 patients, ALN pCR accounted for 46.6

Diabetes mellitus (DM) is associated with systemic metabolic disturbances across multiple organs. Total-body 18F-fluorodeoxyglucose (F-FDG) PET/CT enables simultaneous quantification of glucose metabolism in numerous organs. This study aimed to characterize multi-organ 18F-FDG uptake patterns in type 2 DM patients compared with healthy controls and to explore associations with clinical variables including brain volume. Compared with controls, DM patients exhibited significantly lower SULmean in brain (− 15.3

Background: Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by motor and non-motor symptoms, yet its stage-dependent neurobiological mechanisms remain incompletely understood. Multimodal magnetic resonance imaging (MRI) offers a noninvasive approach to investigate both functional and structural alterations across disease stages. Therefore, this study aimed to characterize stage-dependent functional and iron-related brain alterations in PD using multimodal MRI and to explore their associations with motor severity. Methods: We enrolled 104 PD patients, stratified into early-stage (n=49) and advanced-stage (n=55) based on Hoehn and Yahr (H&Y) score, along with 53 age- and sex- matched healthy controls. Quantitative susceptibility mapping (QSM) quantified iron deposition in the substantia nigra (SN) and globus pallidus (GP), while resting-state functional magnetic resonance imaging (rs-fMRI) was used to assess functional alterations using regional homogeneity (ReHo) and fractional amplitude of low-frequency fluctuations (fALFF). Group comparisons were conducted using one-way analysis of variance (ANOVA) with post-hoc tests, and voxel-wise analyses were corrected using cluster-level false discovery rate (FDR, P<0.05). Correlation analyses were performed to evaluate associations between imaging metrics and Unified Parkinson's Disease Rating Scale part III (UPDRS-III) motor scores. Results: Compared with healthy controls, both early and advanced-stage PD patients showed significantly increased iron deposition in the bilateral SN and GP (all P<0.05), with higher QSM values in advanced-stage PD than in early-stage PD (all P<0.01). Iron deposition in these regions was positively correlated with motor severity assessed by UPDRS-III scores (all P<0.001). Early-stage PD primarily exhibited abnormal fALFF and ReHo in visual-related regions, whereas advanced-stage PD showed more widespread involvement of the basal ganglia-thalamocortical motor circuit, frontoparietal regions, and limbic structures. Functional alterations in motor-related regions were significantly associated with UPDRS-III scores (all P<0.001), while ReHo changes in limbic regions were correlated with cognitive performance (Mini-Mental State Examination, MMSE; P<0.001). Conclusions: Building on established evidence that PD involves progressive iron deposition in the SN and GP and widespread neural network dysfunction, our multimodal MRI findings demonstrate that integrating ReHo, fALFF, and QSM provides a framework for characterizing stage-specific pathophysiological changes and support their potential as biomarkers for early diagnosis, disease staging, and therapeutic development.

RATIONALE AND OBJECTIVES:To investigate the value of time-dependent diffusion MRI (Td-dMRI) and macromolecular proton fraction (MPF) imaging in assessing the pathological grade of cervical cancer (CC). MATERIALS AND METHODS:A total of 92 CC patients, comprising 35 with high-grade (grade III) cancer and 57 with low-grade (grade I/II) cancer, who underwent Td-dMRI and MPF, were prospectively enrolled. Td-dMRI derived parameters including cellularity (cell density), diameter (tumor cell size), Dex (extracellular diffusivity), Vin (intracellular volume fraction), and three apparent diffusion coefficients (ADCPGSE, ADC17 Hz, ADC33Hz) and MPF derived parameter MPF (tissue macromolecular) were calculated and compared. Diagnostic performance was assessed via area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and decision curve analysis (DCA); internal validation was performed using 1000 bootstrap resamples to mitigate model optimism. Multiple pairwise AUC comparisons were adjusted using the DeLong test. RESULTS:Cellularity, Vin, and MPF were higher and diameter, Dex, ADCPGSE, ADC17 Hz, and ADC33 Hz were lower in high-grade group than in low-grade group (all P < 0.05). Cellularity, Vin, and MPF were independent predictors and their combination achieved optimal diagnostic efficacy (AUC = 0.960; 95% CI: 0.897-0.990; sensitivity = 97.14%; specificity = 82.46%), which was significantly higher than any individual parameter (AUC range: 0.685-0.923; all P < 0.05 after Bonferroni correction). Internal validation confirmed stable performance (AUC = 0.955; 95% CI: 0.939-0.960), and DCA demonstrated higher net benefit for patients. Vin strongly correlated with pathological nuclear fraction (r = 0.752, P < 0.001). CONCLUSION:Td-dMRI and MPF were effective methods of predicting pathological grade in CC, and the combination of cellularity, Vin, and MPF has the potential to serve as a new imaging marker, facilitating preoperative grading and personalized treatment.

Understanding the mechanisms limiting OX40 agonist antibody efficacy is critical for developing more effective combination immunotherapies. Tumor microenvironment (TME) analysis revealed that OX40-antibody-responsive mice harbored tumor-associated macrophages (TAMs) with elevated NOS2 expression and heightened pattern recognition receptor (PRR) activation and interferon gamma (IFN-γ) signaling. In addition, patients with more favorable treatment responses to OX40 antibody therapy exhibited increased NOS2 expression. Mechanistically, tumor-infiltrating T-cell-derived IFN-γ synergizes with endogenous ligands of PRR released during immunogenic cell death to drive NOS2+ TAMs reprogramming. Translating these insights into therapeutic strategy, a Combo approach composing of MPLA, IFN-γ, and OX40 agonist antibody is designed to actively polarize TAMs to express NOS2, which mediate tumor clearance through an NOS2-dependent cytotoxicity. Moreover, OX40-antibody-mediated regulatory T cell (Treg) depletion potentiated NOS2+ macrophage induction. This multimodal strategy offers a promising solution to overcome the limitations of OX40 antibody monotherapy and enhance outcomes of the OX40-targeted immunotherapies.

BACKGROUND:Dynamic positron emission tomography (PET) is a powerful tool for clinical tumor diagnosis. However, the conventional dynamic scanning duration takes about 60 m i n $min$ , which is inconvenient for patients and limits the widespread application of this technology. PURPOSE:This study aims to develop an innovative method to achieve shortened high-quality K i $K_i$ parametric imaging based on the Patlak model. METHODS:We proposed a population-based input function integral estimation (PBIF-IE) method. The core of this method is to construct a linear regression model between the early-stage integral ( S e a r l y $S_{early}$ ) and the late-stage mean ( M l a t e $M_{late}$ ) of the image-derived input functions (IDIFs) in the training dataset. The goal is to estimate S e a r l y $S_{early}$ in the testing datasets using M l a t e $M_{late}$ and the linear regression model when only late-stage dynamic sequences are available. To verify the effectiveness and stability of the model, we set up three testing datasets, A, B, and C, each with different framing protocols for K i $K_i$ parametric imaging analysis. Furthermore, to explore the potential of our proposed method in shortening scan duration, we evaluated the K i $K_i$ parameter results obtained by our proposed method under three scan durations (30, 20, and 10 m i n $min$ ) using multiple quantitative metrics, including the peak signal-to-noise ratio (PSNR), the structural similarity index (SSIM), and the relative error (RE). RESULTS:Through multi-center data studies, we demonstrate the effectiveness of the PBIF-IE method. Extensive experimental results demonstrate that the PBIF-IE method outperforms other methods in both S e a r l y $S_{early}$ estimation and K i $K_i$ parametric imaging. The 30 m i n $min$ dynamic scanning protocol can obtain K i $K_i$ parameter images that are highly consistent with the 60 m i n $min$ scanning protocol, while the 20 m i n $min$ dynamic scanning protocol is sufficient for preliminary tumor localization. CONCLUSIONS:Based on the experimental results, the PBIF-IE method outperforms other existing methods for shortened K i $K_i$ parametric imaging. In future research, we plan to explore how the number of dynamic sequences used in the training dataset affects the model construction. This will help us further optimize the parametric imaging process.

OBJECTIVE:Iron deposition is thought to be associated with the physiological mechanisms of Parkinson's disease (PD). However, the pathogenesis remains unclear, and the aim of this study was to investigate the relationship between the progression of iron deposition in the substantia nigra (SN) at different stages of PD and alterations in brain functional connectivity (FC). METHODS:MRI was performed for 106 patients with PD, who were divided into groups with Early Stage (ES) (51 patients) and Advanced Stage (AS) (55 patients) PD according to the Hoehn and Yahr (H&Y) score of clinical severity, along with 45 age and sex matched healthy controls (HC). The iron content of the SN was measured using Quantitative Susceptibility Mapping (QSM), and whole-brain functional connectivity (FC) analysis of functional Magnetic Resonance Imaging (rsfMRI) data was performed using the SN as a seed point. RESULTS:The groups of patients with ES and AS both had significantly increased iron concentration in SN compared to HC, and significantly greater iron in the AS group compared to the ES group. In addition, seed-based FC analysis showed significantly lower FC between the SN and major functional networks, including the sensorimotor network (SMN), the default mode network (DMN), the visual network (VN), the dorsal attentional network (DAN), and the frontal-parietal network (FPN) in the ES and AS compared with the HC group. DISCUSSION:These findings suggest that abnormal iron accumulation in the SN contributes to progressive disruption of brain networks, reflecting the underlying neurodegenerative process of PD. The association between increased iron load and decreased FC highlights the potential role of iron metabolism in driving network-level dysfunction. This provides new insights into the mechanisms linking microstructural pathology to functional disintegration in PD. CONCLUSION:The novelty in this work stems from the changing FC relationships with iron content in the SN and clinical symptoms for the early versus late stage PD subjects. The progression of PD and increase in iron content of the SN, along with the decrease in whole brain FC and increase in clinical symptoms, could provide stage-specific imaging biomarkers for monitoring PD progression. Furthermore, the identified relationship between iron deposition and network dysfunction may inform the future development of network-targeted therapeutic strategies.

To investigate the association between glymphatic function and dopaminergic degeneration in PD assessed via diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) and dopamine transporter imaging striatal binding ratio (DAT-SBR), aiming to clarify their controversial relationship and distinct roles in disease progression. A total of 70 early-stage, drug-naïve patients with PD and 70 age- and sex-matched healthy controls (HCs) were selected from the Parkinson's Progression Markers Initiative database for cross-sectional analysis. Longitudinal data at 4-year follow-up were available for the PD group. Glymphatic function was evaluated using DTI-ALPS, and dopaminergic function using DAT-SBR derived from DAT-SPECT imaging. Clinical motor and non-motor assessments were performed at baseline and follow-up. Correlations between imaging index and clinical variables were analyzed using Spearman correlation and multivariate regression. At baseline, both DTI-ALPS and DAT-SBR index were significantly lower in PD patients compared to HCs. Notably, no significant correlation was observed between ALPS and SBR index. Clinically, the DTI-ALPS index showed negative correlations with body mass index, disease duration, Hoehn and Yahr stage, and UPDRS III scores, and its longitudinal decline correlated with white matter microstructural degeneration. The DAT-SBR index was negatively correlated with Epworth Sleepiness Scale, REM sleep behavior disorder score, and serum urate. Our findings suggest that glymphatic dysfunction and nigrostriatal denervation represent independent, parallel pathological trajectories in early PD. While the ALPS index may serve as a potential imaging marker of structural network integrity linked to motor execution. These indices offer distinct, complementary mechanistic insights into PD pathology.

Metabolic reprogramming is a fundamental hallmark of cancer progression. However, the oncogenic mechanisms underlying serine metabolism and its impact on chemotherapeutic sensitivity in gastric cancer (GC) remain poorly defined. Here, through integrated metabolomics and 13C-labeled metabolic flux analysis, we identify marked dysregulation of serine metabolism in GC, primarily driven by increased expression of phosphoglycerate dehydrogenase (PHGDH). Mechanistically, we show that with no lysine kinase 1 (WNK1) phosphorylates PHGDH at Ser349 and Ser371, enhancing its enzymatic activity and protein stability by preventing ubiquitin-mediated degradation. In vivo, WNK1 knockout mice exhibit significantly reduced gastric tumor burden, accompanied by decreased serine levels and disrupted redox balance, supporting the protumorigenic role of the WNK1-PHGDH axis. Clinically, enhanced PHGDH activity, elevated serine levels, and increased glutathione abundance are strongly associated with poor oxaliplatin response in GC patient cohorts, suggesting PHGDH as a potential predictive biomarker for chemotherapy resistance. Together, these findings delineate a WNK1-PHGDH-driven serine metabolic reprogramming axis that promotes redox adaptation and chemoresistance in GC, highlighting its dual value as a mechanistic driver and a therapeutic vulnerability in cancer treatment.

Parkinson's disease (PD) is an incurable neurological disorder that often begins insidiously with sleep disturbances and somatic symptoms, progressing to whole-body motor and cognitive symptoms1-5. Dysfunction of the somato-cognitive action network (SCAN)-which is thought to control action execution6,7 by coordinating arousal, organ physiology and whole-body motor plans with behavioural motivation-is a potential contributor to the diverse clinical manifestations of PD. To investigate the role of the SCAN in PD pathophysiology and treatments (medications, deep-brain stimulation (DBS), transcranial magnetic stimulation (TMS) and MRI-guided focused ultrasound stimulation (MRgFUS)), we built a large (n = 863), multimodal, multi-intervention clinical imaging dataset. Resting-state functional connectivity revealed that the substantia nigra and all PD DBS targets (subthalamic nucleus, globus pallidus and ventral intermediate thalamus) are selectively connected to the SCAN rather than to effector-specific motor regions. Importantly, PD was characterized by specific hyperconnectivity between the SCAN and the subcortex. We therefore followed six PD cohorts undergoing DBS, TMS, MRgFUS and levodopa therapy using precision resting-state functional connectivity and electrocorticography recording. Efficacious treatments reduced SCAN-to-subcortex hyperconnectivity. Targeting the SCAN instead of effector regions doubled the efficacy of TMS treatments. Focused ultrasound treatment benefits increased when the target was closer to the thalamic SCAN sweet spot. Thus, SCAN hyperconnectivity is central to PD pathophysiology and its alleviation is a hallmark of successful neuromodulation. Targeting functionally defined subcortical SCAN nodes may improve existing therapies (DBS, MRgFUS), whereas cortical SCAN targets offer effective non-invasive or minimally invasive neuromodulation for PD.

Genes impacting the bioaccumulation of perfluoroalkyl and polyfluoroalkyl substances (PFASs)and their neurotoxic effects on the brain and behavior remain unclear. Here,we examined genome-wide associations with serum accumulation of 13 PFASs in 6,823 Chinese adults. We revealed that perfluoroheptanoic acid (PFHpA) accumulation was associated with genetic variants at two loci (3q29: P = 5.20 ×10-19; 6p22.2: P = 3.69 ×10-23), mapping to 56 genes.Blood expression of 27 of these genes was associated with PFHpA accumulation in 573 subsamples. Eight genes showed potential causal effects on PFHpA accumulation,functionally linked to innate immunity (TRIM38, ZDHHC19, MUC20)and organic solute transport (SLC51A and SLC17A3). We assessed the impact of PFASs on cortical thickness and surface area, white matter fractional anisotropy and mean diffusivity,along with 25 behavioral phenotypes. We identified that seven PFASs were correlated with reduced cortical morphology, primarily in the prefrontal cortex. We also found a statistical causal effect of PFHpA accumulation on the surface area in the right frontomarginal cortex, which mediated the effect of PFHpA on anxiety. These findings indicate that serum PFHpA accumulation may be regulated by genes related to innate immunity and solute transport, heightening anxiety by impairing the prefrontal cortex.

Background:Early diagnosis of Parkinson's disease (PD) is crucial for prompt treatment and improved clinical outcomes, but accurate early diagnosis remains challenging. Deep learning (DL) methods have demonstrated significant potential for diagnosing neurological diseases using magnetic resonance imaging (MRI). However, most existing models use generic pattern recognition approaches without incorporating the distinctive characteristics of neuroimaging data and domain-specific knowledge. These limitations restrict both diagnostic accuracy (ACC) and clinical interpretability. This diagnostic ACC study aims to develop a specialized DL framework that effectively integrates neuroimaging domain knowledge to enhance both diagnostic performance and clinical interpretability for early PD detection. Methods:In this study, we propose a Structural and Statistical Knowledge-Enhanced Attention Network (SSKEA-Net) for early PD diagnosis, consisting of two innovative cascaded modules: the Gray-White Interactive Modulation (GWIM) module, which utilizes a structurally specific gray-white matter separation mechanism to modulate channel-wise attention and further enhances tissue-specific features; and the Statistical Prior-Guided Attention (SPGA) module, which incorporates voxel-level statistically significant maps as spatial attention weights to guide feature extraction toward disease-related brain regions. By incorporating the domain knowledge-guided explicit feature enhancement strategy, SSKEA-Net effectively reduces feature interference between gray and white matter, significantly enhancing both model performance and interpretability. We evaluated the diagnostic performance of model using diffusion tensor imaging as input on a rigorously matched early-stage PD dataset with strictly controlled age and gender distributions, and employed activation heatmaps to visualize the interpretability of model. Results:SSKEA-Net achieved superior diagnostic performance in five-fold cross-validation, with an ACC of 0.8798±0.0158, positive predictive value of 0.9185±0.0138, true positive rate of 0.8067±0.0267, specificity of 0.9406±0.0091, and an area under the curve of 0.9301±0.0146, outperforming both classic three-dimensional DL models and the current top-performing neuroimaging model, Simple Fully Convolutional Network. Compared to the baseline model, the cumulative activation heatmaps demonstrated that SSKEA-Net achieved precise anatomical localization, with activation specifically concentrated on the substantia nigra, putamen, midbrain, and corpus callosum, which correspond to clinically relevant brain structures associated with early PD, confirming the enhanced interpretability and clinical relevance. Conclusions:The proposed SSKEA-Net effectively combines domain-specific structural and statistical knowledge with DL, achieving both high diagnostic ACC and clinical interpretability for early PD detection. By incorporating neuroimaging priors into the network architecture, this approach provides a valuable framework for developing more reliable and interpretable artificial intelligence systems in clinical neuroimaging applications.