Xin Meng

Dr. Xin Meng

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 December 2024 in Cancers
Verified
Journal Article
Open access
PDF file
LPLucy Pu
Lucy Pu
RDRajeev Dhupar
Rajeev Dhupar
XMXin Meng
Xin Meng
Background: Surgical resection remains the standard treatment for early-stage lung cancer. However, the recurrence rate after surgery is unacceptably high, ranging from 30% to 50%. Despite extensive efforts, accurately predicting the likelihood and timing of recurrence remains a significant challenge. This study aims to predict postoperative recurrence by identifying novel image biomarkers from preoperative chest CT scans. Methods: A cohort of 309 patients was selected from 512 non-small-cell lung cancer patients who underwent lung resection. Cox proportional hazards regression analysis was employed to identify risk factors associated with recurrence and was compared with machine learning (ML) methods for predictive performance. The goal is to improve the ability to predict the risk and time of recurrence in seemingly “cured” patients, enabling personalized surveillance strategies to minimize lung cancer recurrence. Results: The Cox hazards analyses identified surgical procedure, TNM staging, lymph node involvement, body composition, and tumor characteristics as significant determinants of recurrence risk, both for local/regional and distant recurrence, as well as recurrence-free survival (RFS) and overall survival (OS) (p < 0.05). ML models and Cox models exhibited comparable predictive performance, with an area under the receiver operative characteristic (ROC) curve (AUC) ranging from 0.75 to 0.77. Conclusions: These promising findings demonstrate the feasibility of predicting postoperative lung cancer recurrence and survival time using preoperative chest CT scans. However, further validation using larger, multisite cohort is necessary to ensure robustness and facilitate integration into clinical practice for improved cancer management.

6 February 2025 in Bioengineering
Verified
Journal Article
Open access
PDF file
LPLucas Pu
Lucas Pu
OBOliver Beale
Oliver Beale
XMXin Meng
Xin Meng
Background: Diabetic retinopathy (DR) is the leading cause of blindness among working-age adults. Early detection is crucial to reducing DR-related vision loss risk but is fraught with challenges. Manual detection is labor-intensive and often misses tiny DR lesions, necessitating automated detection. Objective: We aimed to develop and validate an annotation-free deep learning strategy for the automatic detection of exudates and bleeding spots on color fundus photography (CFP) images and ultrawide field (UWF) retinal images. Materials and Methods: Three cohorts were created: two CFP cohorts (Kaggle-CFP and E-Ophtha) and one UWF cohort. Kaggle-CFP was used for algorithm development, while E-Ophtha, with manually annotated DR-related lesions, served as the independent test set. For additional independent testing, 50 DR-positive cases from both the Kaggle-CFP and UWF cohorts were manually outlined for bleeding and exudate spots. The remaining cases were used for algorithm training. A multiscale contrast-based shape descriptor transformed DR-verified retinal images into contrast fields. High-contrast regions were identified, and local image patches from abnormal and normal areas were extracted to train a U-Net model. Model performance was evaluated using sensitivity and false positive rates based on manual annotations in the independent test sets. Results: Our trained model on the independent CFP cohort achieved high sensitivities for detecting and segmenting DR lesions: microaneurysms (91.5%, 9.04 false positives per image), hemorrhages (92.6%, 2.26 false positives per image), hard exudates (92.3%, 7.72 false positives per image), and soft exudates (90.7%, 0.18 false positives per image). For UWF images, the model’s performance varied by lesion size. Bleeding detection sensitivity increased with lesion size, from 41.9% (6.48 false positives per image) for the smallest spots to 93.4% (5.80 false positives per image) for the largest. Exudate detection showed high sensitivity across all sizes, ranging from 86.9% (24.94 false positives per image) to 96.2% (6.40 false positives per image), though false positive rates were higher for smaller lesions. Conclusions: Our experiments demonstrate the feasibility of training a deep learning neural network for detecting and segmenting DR-related lesions without relying on their manual annotations.

28 January 2026 in Bioengineering
Verified
Journal Article
Open access
PDF file
LPLucy Pu
Lucy Pu
NGNaciye Sinem Gezer
Naciye Sinem Gezer
TYTong Yu
Tong Yu
ZKZehavit Kirshenboim
Zehavit Kirshenboim
EDEMRAH DUMAN
EMRAH DUMAN
XMXin Meng
Xin Meng
Mucus plugs are airway-obstructing accumulations of inspissated secretions frequently observed in obstructive lung diseases (OLDs), including chronic obstructive pulmonary disease (COPD), severe asthma, and cystic fibrosis. Their presence on chest CT is strongly associated with airflow limitation, reduced lung function, and increased mortality, making them emerging imaging biomarkers of disease burden and treatment response. However, manual delineation of mucus plugs is labor-intensive, subjective, and impractical for large cohorts, leading most prior studies to rely on coarse segment-level scoring systems that overlook lesion-level characteristics such as size, extent, and location. Automated plug-level quantification remains challenging due to substantial heterogeneity in plug morphology, overlap in attenuation with adjacent vessels and airway walls on non-contrast CT, and pronounced size imbalance in clinical datasets, which are typically dominated by small distal plugs. To address these challenges, we developed and validated a simulation-driven, annotation-free deep learning framework for automated detection and segmentation of airway mucus plugs on non-contrast chest CT. A total of 200 COPD CT scans were analyzed (98 plug-positive, 83 plug-negative, and 19 uncertain). Synthetic mucus plugs were generated within segmented airways by transferring voxel-intensity statistics from adjacent intrapulmonary vessels, preserving realistic morphology and texture while enabling controlled sampling of plug phenotypes. An nnU-Net trained exclusively on synthetic data (S-Model) was evaluated on an independent, expert-annotated test set and compared with an nnU-Net trained on manual annotations using 10-fold cross-validation (M-Model). The S-Model achieved significantly higher detection performance than the M-Model (sensitivity 0.837 [95% CI: 0.818–0.854] vs. 0.757 [95% CI: 0.737–0.776]; 1.91 false positives per scan vs. 3.68; p < 0.001), with performance gains most pronounced for medium-to-large plugs (≥6 mm). This simulation-driven approach enables accurate, scalable quantification of mucus plugs without voxel-wise manual annotation in thin-slice (<1.5 mm) non-contrast chest CT scans. While the framework could, in principle, be extended to other annotation-limited medical imaging tasks, its generalizability beyond this COPD cohort and imaging protocol has not yet been established, and future work is required to validate performance across diverse populations and scanning conditions.


26 September 2024 in Cancers
Verified
Journal Article
Open access
PDF file
TYTong Yu
Tong Yu
XZXiaoyan Zhao
Xiaoyan Zhao
JLJoseph K. Leader
Joseph K. Leader
JWJing Wang
Jing Wang
XMXin Meng
Xin Meng
JPJiantao Pu
Jiantao Pu
Objective: This study aims to investigate the association between the arteries and veins surrounding a pulmonary nodule and its malignancy. Methods: A dataset of 146 subjects from a LDCT lung cancer screening program was used in this study. AI algorithms were used to automatically segment and quantify nodules and their surrounding macro-vasculature. The macro-vasculature was differentiated into arteries and veins. Vessel branch count, volume, and tortuosity were quantified for arteries and veins at different distances from the nodule surface. Univariate and multivariate logistic regression (LR) analyses were performed, with a special emphasis on the nodules with diameters ranging from 8 to 20 mm. ROC-AUC was used to assess the performance based on the k-fold cross-validation method. Average feature importance was evaluated in several machine learning models. Results: The LR models using macro-vasculature features achieved an AUC of 0.78 (95% CI: 0.71–0.86) for all nodules and an AUC of 0.67 (95% CI: 0.54–0.80) for nodules between 8–20 mm. Models including macro-vasculature features, demographics, and CT-derived nodule features yielded an AUC of 0.91 (95% CI: 0.87–0.96) for all nodules and an AUC of 0.82 (95% CI: 0.71–0.92) for nodules between 8–20 mm. In terms of feature importance, arteries within 5.0 mm from the nodule surface were the highest-ranked among macro-vasculature features and retained their significance even with the inclusion of demographics and CT-derived nodule features. Conclusions: Arteries within 5.0 mm from the nodule surface emerged as a potential biomarker for effectively discriminating between malignant and benign nodules.

1 September 2026 in Journal of Medical Imaging
Verified
Journal Article
Closed access
JWJing Wang
Jing Wang
TYTong Yu
Tong Yu
HLHao-En Lu
Hao-En Lu
ZZZixue Zeng
Zixue Zeng
XMXin Meng
Xin Meng
JPJiantao Pu
Jiantao Pu
PurposeTo evaluate the feasibility and challenges of heart chamber segmentation from noncontrast CT scans using contrastive unpaired image translation and deep learning-based segmentation.ApproachWe developed ChameleonNet, a framework utilizing the Contrastive Unpaired Translation (CUT) network with decoupled contrastive learning (DCL) loss to synthesize noncontrast CT from contrast CT scans. Using annotations of four heart chambers (left atrium (LA), left ventricle (LV), right atrium (RA), and right ventricle (RV)) from contrast scans, we trained a Hausdorff distance loss-enhanced nnU-Net on synthesized noncontrast images. The translation model was trained with 35,538 contrast-enhanced and 37,197 noncontrast CT slices. The segmentation model was trained with 292 synthesized noncontrast scans. Performance was evaluated using Dice similarity coefficient (DSC) and 95th Hausdorff distance (HD95) on 36 synthesized noncontrast scans, and volume agreement on 36 real noncontrast CT scans was assessed using Pearson correlation, mean absolute percentage error (MAPE), and mean percentage error (MPE).ResultsThe segmentation model achieved DSC of 0.94 (±0.01), 0.91 (±0.04), 0.92 (±0.03), 0.93 (±0.02), and HD95 of 3.63 (±1.49), 5.74 (±4.08), 5.18 (±1.77), and 5.51 (±3.21) mm on synthesized noncontrast images for LA, LV, RA, and RV, respectively. On real noncontrast CT scans, Pearson correlations were 0.93, 0.82, 0.87, and 0.89 (all p<0.001), with MAPE ranging from 9.22% to 20.79%, and MPE ranging from −12.52% to 4.67%.ConclusionsChameleonNet demonstrated feasibility for heart chamber segmentation from noncontrast CT without manual noncontrast annotations. However, volume errors, particularly for LV and RV, indicate that further refinement and validation are needed before clinical use.