BPBalaji Pandian

Balaji Pandian

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

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


Balaji Pandian has a total of 57 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Machine Learning to Identify Point-of-Care Ultrasound and Evaluate Standardized Documentation: Retrospective Operational Cohort Study

    by Kevin Nguyen, Zewen Wu, Chu-An Tsai, John Vandervest, D’ANna Lammers, Ruth Cassidy, Zachary Murphy, Balaji Pandian, Maya M Hammoud, Jennifer Collin, et al.

    Background: Point-of-care ultrasound (POCUS) is integral to obstetrics and gynecology (OBGYN), offering bedside diagnostic and therapeutic advantages. Despite its widespread adoption, accurate documentation and billing remain challenging due to inconsistent workflows, variable free-text note quality, and inefficiencies within electronic health record (EHR) systems. These barriers often result in missed procedural charges and hinder operational, educational, and reimbursement efforts. Objective: This study leveraged machine learning (ML) to automatically identify POCUS procedures within clinical notes and assessed the effect of implementing standardized procedure documentation (ProcDoc) templates on billing capture accuracy and efficiency. Methods: We conducted a multipart retrospective cohort study at a large academic medical center using EHRs from January 2018 to August 2024 across 11 OBGYN clinic sites. ML models (LightGBM [light gradient boosting machine] and BioClinBERT [biomedical and clinical bidirectional encoder representations from transformers]) were trained on clinical encounter notes to classify POCUS procedures and validated against Current Procedural Terminology (CPT) manual code assignments. In February 2023, a standardized ProcDoc smart form was introduced to streamline POCUS documentation and automatically trigger CPT billing codes. Preintervention and postintervention periods were compared using ML metrics and manual billing audits. Outcomes included model accuracy, recall, precision, adoption rates, improvement in billing recapture, and usage of ProcDoc templates. Results: A total of 559,029 encounters from 109,776 unique patients were analyzed. The BioClinBERT model (accuracy 0.97; F 1 -score 0.55-0.63) demonstrated a robust ability to identify documented and missed procedures in free-text clinical notes. ProcDoc adoption reached 75.1% within 12 months, supported by comprehensive staff education. Billing recapture—the proportion of charges missed by providers but later identified—dropped from 10.0% preintervention to 2.4% postintervention, primarily arising from the shift toward auto-capturing documentation (odds ratio 0.22, 95% CI 0.17‐0.30; P <.001), with overall POCUS billing slightly increased (+0.6%). Most postintervention CPT codes (1812/2404, 75.4%) originated from ProcDoc templates, confirming improved workflow efficiency and reduced manual audit burden. Model analysis and billing metrics demonstrated that improvements were associated with workflow changes and not an increase in procedure frequency. Conclusions: ML modeling proved effective for extracting POCUS procedures from clinical documentation and serving as an evaluation tool for workflow interventions. Standardized documentation with ProcDoc significantly enhanced charge capture accuracy and reduced dependence on manual chart reviews and billing reconciliation. This approach highlights the use of ML as a retrospective auditing and evaluation tool for assessing clinical workflow interventions. Broader application of similar strategies could address documentation inefficiencies and promote sustainability across health care settings.

  • Open Access

    Rapid, label-free detection of diffuse glioma recurrence using intraoperative stimulated Raman histology and deep neural networks

    by Todd C Hollon, Balaji Pandian, Esteban Urias, Akshay V Save, Arjun R Adapa, Sudharsan Srinivasan, Neil K Jairath, Zia Farooq, Tamara Marie, Wajd N Al-Holou, et al.

    Detection of glioma recurrence remains a challenge in modern neuro-oncology. Noninvasive radiographic imaging is unable to definitively differentiate true recurrence versus pseudoprogression. Even in biopsied tissue, it can be challenging to differentiate recurrent tumor and treatment effect. We hypothesized that intraoperative stimulated Raman histology (SRH) and deep neural networks can be used to improve the intraoperative detection of glioma recurrence. We used fiber laser–based SRH, a label-free, nonconsumptive, high-resolution microscopy method (2) to image a cohort of patients (n = 35) with suspected recurrent gliomas who underwent biopsy or resection. The SRH images were then used to train a convolutional neural network (CNN) and develop an inference algorithm to detect viable recurrent glioma. Following network training, the performance of the CNN was tested for diagnostic accuracy in a retrospective cohort (n = 48). Using patch-level CNN predictions, the inference algorithm returns a single Bernoulli distribution for the probability of tumor recurrence for each surgical specimen or patient. The external SRH validation dataset consisted of 48 patients (recurrent, 30; pseudoprogression, 18), and we achieved a diagnostic accuracy of 95.8%. SRH with CNN-based diagnosis can be used to improve the intraoperative detection of glioma recurrence in near-real time. Our results provide insight into how optical imaging and computer vision can be combined to augment conventional diagnostic methods and improve the quality of specimen sampling at glioma recurrence.


  • Open Access

    MiPanda: A Resource for Analyzing and Visualizing Next-Generation Sequencing Transcriptomics Data

    by Yashar S. Niknafs, Balaji Pandian, Tilak Gajjar, Zach Gaudette, Kevin Wheelock, Mitra P. Maz, Rohan K. Achar, Melinda Song, Cory Massaro, Xuhong Cao, et al.


  • Posterior Fossa Craniotomy for Adherent Fourth Ventricle Neurocysticercosis

    by Lynze R Franko, Balaji Pandian, Avneesh Gupta, Luis E Savastano, Kevin S Chen, James Riddell, Daniel A Orringer

    BACKGROUND AND IMPORTANCE: Neurocysticercosis (NCC) is an infectious helminthic disease often presenting in patients who have immigration or travel history from areas where NCC is endemic. Fourth ventricle cysts from NCC pose a unique treatment challenge, as there is little consensus on the best treatment. This case study describes the treatment of a patient with fourth ventricle neurocysticercosis (FVNCC), examines the therapeutic decision-making, and provides a video of a posterior fossa craniotomy (PFC) resection of a degenerative cyst. CLINICAL PRESENTATION: The patient presented with headache, dizziness, nausea, and memory difficulties. A fourth ventricle cyst consistent with NCC was found on magnetic resonance imaging, and serum enzyme-linked immunosorbent assay (ELISA) confirmed the diagnosis. The cyst was removed utilizing an open PFC followed by antihelminthic therapy and corticosteroids. There was resolution of symptoms at 9 mo postoperatively. CONCLUSION: Several treatment modalities have been proposed for isolated cysts in the fourth ventricle, including medication, ventriculoperitoneal shunt, endoscopic removal, and PFC. The treatment decision is complex, and there is little guidance on the best treatment choices. In this article, we describe treatment via PFC for an adherent FVNCC cyst.


  • Virtual brain tumor histopathology and immunohistochemistry with stimulated raman scattering microscopy.

    by Spencer B Lewis, Balaji Pandian, Todd Hollon, Yashar Niknafs, Mia Garrard, Sandra Camelo-Piragua, Daniel A Orringer

    E13511 Background: Stimulated Raman scattering (SRS) microscopy is a label-free optical technique which uses the chemical composition of fresh tissue to generate image contrast. Prior work has demonstrated exceptional correlation between SRS images and H&E microscopy for detecting brain tumor infiltration ex vivo. However, SRS has not previously been suitable for a clinical environment. Further, SRS has utilized structural, rather than chemical, diagnostic features. We present the first clinical validation of SRS, as well as our work towards mutation-targeted SRS imaging. Methods: Implementation of SRS with a fiber laser source allowed sufficient stability for clinical deployment. Unprocessed specimens from 200 neurosurgical cases were imaged at the CH2 and CH3 vibrational modes, then processed to mimic the appearance of H&E staining. 30 cases were trialed in a simulated intraoperative consultation to assess concordance between SRS and cryosection microscopy. A multi-layer perceptron (MLP) model was then used to predict brain tumor diagnosis based on quantified image attributes. To explore the potential for chemically specific SRS, isogenic lines of normal human astrocyte were prepared, differing by the presence or absence of the R132H mutation in the isocitrate dehydrogenase 1 (IDH1) gene. IDH1 R132H cells accumulate high concentrations of the oncometabolite 2-hydroxyglutarate and display predictably altered metabolism. 200 Raman spectra were collected from 50 wild type (WT) and 50 R132H cells with the aim of detecting these changes. Each cell line was then submitted for targeted metabolomic analysis. Results: Excellent concordance was observed between SRS and frozen section histology ( κ >0.89). The MLP diagnosed brain tumor subtype with 90% accuracy. A robust spectral difference was observed between wild type and R132H cells, which showed concordance with metabolomic data. Conclusions: The diagnostic capability, ease of use and speed of SRS make it well-suited for integration into the operative workflow. Further, because SRS amplifies the Raman signal with no resonant background, the spectral differences observed may allow rapid differentiation of IDH1 WT and R132H cells with SRS.