AGAjay Gulati

Ajay Gulati

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ajay Gulati has a total of 70 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • ConLymphNet: A Generalizable Region-Constrained Deep Learning Architecture for Precise Abdominal Lymph Node Segmentation Across Cancer Types.

    by Pankaj Gupta, Shubham Saini, Anjali Aggarwal, Ajay Gulati, Shraddha Patkar, Niharika Dutta, Sakshi Jeswani, Lovleen Kakkar, Kartik Bose, Gaurav Prakash, et al.

    Accurate identification and segmentation of abdominal lymph nodes on computed tomography (CT) is crucial for cancer staging and treatment planning but remains a challenging and time-consuming radiologic task. We developed and validated ConLymphNet, a deep learning-based approach for automated abdominal lymph node segmentation using vertebral landmarks as anatomic reference points to standardize the region of interest. We implemented a novel preprocessing pipeline incorporating vertebral landmark-based region selection followed by a 3D nnUNet segmentation model. The model was trained on 481 contrast-enhanced CT scans from patients with gallbladder cancer (GBC) and validated on external data sets including GBC (n = 54), public lymph node (n = 81), and multicancer data sets (n = 606). Performance was compared with three radiologists of varying experience levels. Interobserver agreement among radiologists and the incremental value of artificial intelligence (AI)-assisted reading were also assessed. To support reproducibility and future research, we also contribute 45 expert-verified lymph node segmentations for cases from public multicancer data sets. ConLymphNet achieved mean Dice coefficients of 0.696 ± 0.179, 0.631 ± 0.187, and 0.639 ± 0.176 and low false-positive rates of 0.78 ± 0.35, 0.81 ± 0.33, 0.93 ± 0.47/volume on internal cross-validation, external GBC, and public lymph node data sets, respectively. Performance varied by lymph node location (P = .0080) and size (P = .0444), with larger nodes yielding better results. The model's detection rate (89.6%) was comparable with experienced residents (89.2%) but with reduced processing time (28.7 ± 12.4s v 165.3 ± 47.8 s). AI assistance improved radiologists' performance significantly, particularly for less experienced readers (detection rate +20.2%, segmentation accuracy +13.0%, P < .001), while reducing false positives by 33.3%. ConLymphNet demonstrates reasonable performance across diverse cancer types and imaging protocols, offering segmentation accuracy comparable with radiologists with substantially faster processing times.

  • Artificial Intelligence for Molecular Subtyping in Unresectable Gallbladder Cancer: A Proof-of-Concept Study for CT-based HER2 Status Prediction

    by Pankaj Gupta, Chetan Madan, Niharika Dutta, Shravya Singh, Nikita Pradhan, Ruby Siddiqui, Ajay Tomar, Ajay Gulati, Naveen Kalra, Gaurav Prakash, et al.


  • Magnetic Resonance Elastography Versus Shear Wave Elastography in Chronic Liver Disease

    by Jaswinder Singh, Tarvinder Singh, Naveen Kalra, Ajay Gulati, Harish Bhujade, Ajay Duseja, Ashim Das


  • Open Access

    Multiple instance learning approach for automated gallbladder cancer detection using ultrasound imaging: multi-center validation of a deep learning model

    by Pankaj Gupta, Kartik Bose, Priya Mudgil, Niharika Dutta, Ajay Gulati, Gaurav Prakash, Thakur Deen Yadav, Lileswar Kaman, Santosh Irrinki, Harjeet Singh, et al.


  • Open Access

    CHEK2 germline gene variant in a patient with a somatotrophinoma and primary hyperparathyroidism: a novel MEN1-like syndrome?

    by Durairaj Arjunan, Mohammad Hayat Bhat, Ashutosh Rai, Vaishali Kaur, Sayka Barry, Md Ejaz Alam, Mohammad Salem Baba, Ajay Gulati, Debajyoti Chatterjee, Manzoor Ahmad Latoo, et al.

    Context: Somatotrophinomas can occasionally occur in familial settings and may be associated with known germline mutations, such as MEN1, AIP, CDKN1B, PRKAR1A, SDHx, and MAX. Recently, the CHEK2 gene has emerged as a potential pituitary tumour predisposition gene. Objective: To present a rare case of a patient with a pituitary somatotrophinoma and primary hyperparathyroidism (PHPT) associated with a likely pathogenic CHEK2 germline mutation, suggesting a novel MEN1-like phenotype. Methods: We conducted a detailed clinical, biochemical, radiological, and genetic evaluation of a 38-year-old woman presenting with features of acromegaly and PHPT. Genetic testing for known MEN1 syndrome-associated genes and broader pituitary tumour predisposition genes was performed. Results: The patient presented with features of acromegaly of 3 years duration and a collagenoma. Laboratory evaluation revealed an elevated IGF-1. Biochemical and imaging studies also revealed PTH-dependent hypercalcaemia and bilateral inferior parathyroid adenomas. Genetic testing for a panel of genes causing hypercalcaemia – including MEN1 and CDKN1B – was negative; however, a pathogenic nonsense variant in CHEK2 (c.232C>T; p.(Gln78Ter); gnomAD frequency: 0.0006%) was detected with exome sequencing. The patient underwent transsphenoidal resection of the pituitary tumour followed by gamma-knife radiosurgery and received long-acting octreotide every four weeks. Parathyroidectomy was performed 1.5 years later. Conclusion: This is the first reported case of a CHEK2 germline mutation associated with somatotrophinoma, PHPT, and collagenoma, mimicking the clinical MEN1 syndrome. These findings expand the spectrum of possible CHEK2-associated neoplasia and highlight the need to consider CHEK2 as a possible candidate gene in patients with MEN1-like syndromes when common mutations have been excluded.