Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Justin Ryan Lay Tan has a total of 46 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Dual-Encoder Network Optimization for Binary Semantic Segmentation of Focal Liver Lesions (FLLs)

    by Jessica Velasco, Melvin Cabatuan, Argel Bandala, Edwin Sybingco, Cesar Llorente, Rennan Baldovino, Laurence Gan Lim, James Manuel Medalla, Stephen Wong, Justin Ryan Tan

    Automated binary semantic segmentation of focal liver lesions (FLLs) in multi-phasic computed tomography (CT) scans is important for knowledge extraction of hepatic malignancy, staging disease, and planning treatment. However, standard models struggle due to extreme class imbalances, complex boundary transitions, and tissue heterogeneity across phases. We present a systematic ablation study evaluating CNN Only, Transformer Only, and parallel dual-encoder networks that combine a local convolutional neural network (CNN) branch (VGG-19 or ConvNeXt-Tiny) with a global hierarchical vision transformer branch (Swin or CSwin) against standard and state-of-the-art baselines for semantic segmentation in medical images. These backbones were evaluated across feature fusions as: early, intermediate, and late strategies on a 308-patient subset of the MCT-LTDiag dataset, equally partitioned across five FLL subtypes (BCLM, CRLM, HCC, HH, and ICC). The study evaluates binary lesion segmentation (Foreground vs. Background Liver) across 5 distinct clinical subtype cohorts (BCLM, CRLM, HCC, HH, and ICC). Evaluation was conducted via patient-grouped, stratified three-fold cross-validation. Pairwise architectural comparisons were performed using a patient-level paired design, with statistical significance evaluated via bootstrap 95% confidence intervals (10,000 resamples), two-sided paired permutation tests (10,000 sign-flips), and Cohen’s du effect sizes adjusted using the Holm-Bonferroni method. The proposed dual-encoder configuration utilizing ConvNeXt-Tiny + Swin Transformer with intermediate feature fusion achieved an optimal trade-off in segmentation performance, yielding a top Mean Dice score of 0.875 ± 0.076 and a Mean IoU of 0.861 ± 0.082. Compared to the best CNN-only baseline (ConvNeXt-Tiny), the proposed model showed statistically significant improvements across all parameters: +0.032 Mean Dice, −15.00 mm HD95, and +0.137 Normalized Surface Dice (NSD). From a resource-complexity perspective, the optimal ConvNeXt-Tiny + Swin Transformer with Intermediate Fusion model operated at <8.1% of the GFLOPs (33.34 vs. 412.30 GFLOPs) and achieved ~6.7 times faster with inference speedup (24.66 ms vs. 165.00 ms latency) than computationally intensive state-of-the-art 2D nnU-Net baselines.

  • Open Access

    Dual-Encoder Network Optimization for Binary Semantic Segmentation of Focal Liver Lesions (FLLs)

    by Jessica Velasco, Melvin Cabatuan, Argel A. Bandala, Edwin Sybingco, Cesar Llorente, Rennan Baldovino, Laurence Gan-Lim, James Manuel Medalla, Stephen Wong, Justin Ryan Tan

    Automation of semantic segmentation of binary detection of focal liver lesions (FLLs) in triphasic computed tomography (CT) scans is critical for knowledge extraction of hepatic malignancy, staging a disease, and planning a treatment. But standard models struggle with class imbalances, boundary differences, and tissue heterogeneity in different phases. This study proposes a modified architecture of a parallel dual-encoder network for FLL boundary refinement optimization. The proposed architecture implements an early feature fusion by stacking the non-contrast (NC), arterial (ART), and portal venous (PVP) phases in the channel dimension. Alongside a per scale, lightweight fusion strategy integrating localized convolutional details to global transformer context. The two architectures were modified and optimized: VGG-19 paired with a Swin Transformer and ConvNeXT-Small paired with a Cross-Shape Window (CSwin) Transformer. The study used a patient-grouped stratified 3-fold validation on 517-case MCT-LTDiag dataset. The networks were extensively benchmarked against standard U-Net, DECTNet and nnU-Net baselines across multiple performance metrics: generalization, segmentation, boundary, and computational. Statistical significance was precisely tested by using a two-tailed paired t-test with Benjamini-Hochberg False Discovery Rate (BH-FDR) adjustments. The ConvNeXT-Small + CSwin Transformer configurations turned out to be the best architecture, with an elite average Dice score of 0.9091 and a leading Intersection over Union (IoU) of 0.8924. This model reduced boundary errors and achieved the lowest 95th percentile Hausdorff Distance (HD95) of 13.86. It resolved the precision-sensitivity trade-off plaguing the baseline models by maintaining a leading sensitivity of 0.9484 and a precision of 0.9352. The champion architecture sustained a tight footprint of ~29M parameters and achieved an optimized average inference speed of 0.111 seconds per slice, even though it has an advanced attention mechanism. This cut the process delay of the VGG-19 variant in half. Qualitative and quantitative results confirm the synergy of modern localized depthwise convolutions and global cross-shaped attention mechanisms removed the bloated false-positive masks and broken under-segmentation. It is delivering a highly stable, precise, and reproducible tool for automated clinical workflows.

  • Open Access

    Comparison of colon adenoma detection rate using cap-assisted and artificial intelligence-assisted colonoscopy at a tertiary hospital in the Philippines: a propensity score-matched analysis

    by Justin Ryan Lay Tan, Keith Brian Tan Enriquez, Kenneth Vergel Tecson Aballe, Mary Anne Gonzales Go, Michael Louie Ong Lim, Jonard Tan Co

    Background/Aims: The integration of artificial intelligence (AI)-powered image analysis and mucosal exposure devices, such as distal attachment caps, has been demonstrated to significantly improve the adenoma detection rate (ADR) during colonoscopy. This study aimed to compare AI-assisted colonoscopy (AIC) with cap-assisted colonoscopy (CAC).Methods: This retrospective propensity score-matched cohort study was performed at a tertiary care hospital between January 2022 and May 2022. Data were extracted from the electronic health record system and colonoscopy video recordings. Adult patients aged 40 years who underwent screening or surveillance colonoscopies were included. The primary outcome was the ADR, whereas the secondary outcome was the polyp detection rate (PDR).Results: A 1:1 propensity score-matched analysis was performed, resulting in 49 well-matched patient pairs. One patient from each pair was assigned to the CAC group, whereas the other was assigned to the AIC group. No significant difference in ADR was observed between the CAC and AIC groups (47% vs. 51%, p=0.69). Similarly, PDR did not significantly differ between the two groups (80% vs. 71%, p=0.35).Conclusions: Both CAC and AIC have the potential to increase ADR and PDR. However, neither modality offers a significant advantage.


  • Open Access

    Successful resection of ampullary neuroendocrine tumors using clip traction-assisted hybrid endoscopic submucosal dissection

    by Ke Jiang, Justin Ryan Lay Tan, Weigang Gu, Qifeng Lou, Hangbin Jin, Jianfeng Yang, Xiaofeng Zhang

    A 63-year-old man was referred due to a 1-month history of abdominal pain. Endoscopic ultrasound (EUS) and endoscopic retrograde cholangiopancreatography (ERCP) showed a 2.1 × 2.3 cm duodenal papilla submucosal tumor (SMT) with distinct borders ([Fig. 1]). Endoscopic submucosal dissection (ESD) was initially planned. A submucosal injection of diluted methylene blue was performed, followed by a mucosal incision using a GoldKnife (Microtech Nanjing). This was followed by a layered multi-knife dissection (Gold Knife and IT Knife Nano) performed with clip traction assistance ([Fig. 2] a, b). Due to anatomical challenges and bleeding, hybrid-ESD was adopted ([Fig. 2] c, d). Placement of pancreatic and biliary stents for endoscopic retrograde pancreatic drainage (ERPD) and endoscopic retrograde biliary drainage (ERBD) and the post-resection defect were closed with a total of seven hemoclips. A nasobiliary tube was inserted to monitor the bleeding ([Fig. 3], [Video 1]). The patient was discharged 7 days post-procedure. The combined maximum diameter of the two specimens measured 1.9 cm ([Fig. 4] a, b). Histopathology and immunohistochemistry showed a well-differentiated neuroendocrine tumor (NET), classified as G1 ([Fig. 4] c, d), with tumor-free resection margins. Follow-up ERCP after 1 month demonstrated complete healing of the excision site, and random biopsy results were normal ([Fig. 5], [Video 1]). The patient remained asymptomatic with no evidence of recurrence at the 12-month follow-up. The unique anatomical features of the duodenum pose certain challenges to the implementation of ESD [1] [2] [3] [4]. In this context, we highlight the utility of a single-clip-assisted traction technique combined with a novel hybrid-ESD approach for treating ampullary lesions. Our experience demonstrates that this technology effectively enhances surgical field exposure, thereby improving dissection efficiency and the overall resection rate. Additionally, it reduces risks such as bleeding and perforation. Endoscopy_UCTN_Code_TTT_1AO_2AG_3AD Article published online: 26 June 2025 © 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/). Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany

  • Open Access

    EUS-guided drainage of multidrug-resistant enterococcal caudate lobe abscess inaccessible with percutaneous approach (with video)

    by Yangyang Zheng, Qiaoxian He, Weigang Gu, Justin Ryan Tan, Jianfeng Yang, Xiaofeng Zhang, Hangbin Jin

  • Open Access

    Directional movement guide for digital single-operator cholangioscopy system

    by Weigang Gu, Justin Ryan L. Tan, Hangbin Jin, Qifeng Lou, Chuang Tang, Ka Shing Cheung, Jianfeng Yang, Xiaofeng Zhang

  • Open Access

    Balloon catheter‐assisted endoscopic resection for papillary adenoma of non‐exposed protruded type (with video)

    by Weigang Gu, Justin Ryan L. Tan, Hangbin Jin, Qifeng Lou, Chuang Tang, Ka Shing Cheung, Jianfeng Yang, Xiaofeng Zhang

    Papillary adenomas, known precursors to papillary adenocarcinoma, warrant close monitoring due to their malignant potential. Historically, surgical resection represented the mainstay of treatment for papillary adenomas with intraductal extension. However, recent advancements in endoscopic techniques have facilitated the adoption of endoscopic papillectomy as a minimally invasive alternative in carefully selected cases. We report a case of an 82‐year‐old woman with a diagnosis of papillary adenoma exhibiting intraductal extension. This was managed with a novel endoscopic technique, balloon catheter‐assisted endoscopic resection. Due to the obscured intraductal component of the papillary mass, a balloon occlusion catheter was deployed within the common bile duct and used as traction to facilitate endoscopic visualization of the mass. Endoscopic resection via papillectomy was subsequently performed. Histopathological examination of the resected specimen revealed a villous adenoma with high‐grade dysplasia. Serial endoscopic ultrasound examinations with targeted papillary biopsies were performed to monitor for disease recurrence.

  • Open Access

    Ischemic Polypectomy Through Detachable Snare and Rubber Band Ligation in Peutz–Jeghers Syndrome

    by Justin Ryan L. Tan, Jonard T. Co

    Endoscopic polypectomy is essential for the prevention of Peutz–Jeghers syndrome–associated complications, including intussusception, intestinal obstruction, and malignant transformation. Conventional polypectomy is the preferred approach, but it can be challenging to achieve in patients with Peutz–Jeghers syndrome because of the high polyp burden and polyps located in areas with difficult endoscopic access. This case report highlights 2 different techniques of ischemic polypectomy and its advantage compared with conventional polypectomy in this subset of patients.


  • Open Access

    Ustekinumab as a Treatment for Ixekizumab-Associated New-Onset Crohn's Disease in a Patient with Psoriasis

    by Justin Ryan L. Tan, Willy S. Alba

    Interleukin-17 inhibitors are effective treatments for plaque psoriasis. However, these medications have been linked to the development of new-onset inflammatory bowel disease (IBD) and the worsening of existing IBD in some patients. This case report describes a patient with plaque psoriasis who developed new-onset Crohn's disease after treatment with ixekizumab, an interleukin-17A inhibitor. He was then transitioned to ustekinumab, which resulted in successful remission of both psoriasis and Crohn's disease. This case highlights the potential for ustekinumab to be an effective rescue treatment for psoriasis patients with new-onset IBD triggered by medications.


  • Open Access

    The characteristic “alveolus nest sign” in solid pseudopapillary neoplasm of the pancreas

    by Weigang Gu, Justin Ryan L. Tan, Hang Bin Jin, Quifeng Lou, Ka Shing Cheung, Jianfeng Yang, Xiaofeng Zhang

    A 23-year-old woman presented with a 1-week history of abdominal pain radiating to the back. The pain was not relieved by proton pump inhibitors. Contrast-enhanced abdominal computed tomography scan demonstrated a round, well-defined, hypodense lesion within the pancreatic body ([Fig. 1]). On endoscopic ultrasound (EUS) elastography there was a 2 × 2.5-cm hypoechoic mass that was predominantly hard (blue) with dispersed heterogeneous soft (green) areas ([Fig. 2]). Contrast-enhanced harmonic endoscopic ultrasound (CH-EUS) revealed hyperechoic solid granular components interspersed with small anechoic regions within the pancreatic mass ([Fig. 3]). These characteristics were typical of the “alveolus nest sign” and were present during the arterial and venous phases of CH-EUS ([Video 1]). EUS-guided fine-needle biopsy with subsequent histopathologic examination ([Fig. 4]) and immunohistochemical analysis ([Fig. 5]) yielded a definitive diagnosis of solid pseudopapillary neoplasm (SPN). The patient underwent distal pancreatectomy for definitive treatment. SPN is an uncommon pancreatic neoplasm exhibiting low malignant potential. It predominantly affects young females and is characterized by indolent behavior, with an excellent long-term prognosis for the majority of patients [1] [2] [3] [4]. CH-EUS examination may reveal characteristic features suggestive of SPN pathology, including the presence of an intralesional pseudopapillary architecture or intratumoral clefting, also termed the “alveolus nest sign” [1] [2]. On CH-EUS, the alveolus nest sign is seen as a region with isoechogenicity to hyperechogenicity, containing a solid and granular component. This component is interspersed with multiple small anechoic areas of varying sizes. These anechoic regions become evident on CH-EUS typically 40 seconds after contrast injection. The presence of this sign at any time within 40 seconds to 5 minutes post-contrast administration is considered characteristic of the alveolus nest sign [1] [2]. CH-EUS is useful in differentiating SPN of the pancreas from other pancreatic tumors. Alveolus nest sign is a characteristic feature of SPN on CH-EUS. Endoscopy_UCTN_Code_CCL_1AF_2AZ Article published online: 29 July 2024 © 2024. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (https://creativecommons.org/licenses/by/4.0/). Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany