Prabhakar Ramachandran

Prof. Prabhakar Ramachandran


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
11 July 2022 in Physical and Engineering Sciences in Medicine
Verified
Journal Article
Open access
PDF file
BBBenjamin Broadhead
Benjamin Broadhead
CNChristopher Noble
Christopher Noble
PRPrabhakar Ramachandran
Prabhakar Ramachandran
Optically stimulated luminescence dosimetry is a relatively recent field of in-vivo dosimetry in clinical radiotherapy, developing over the last 20 years. As a pilot study, this paper presents a direct comparison between the sensitivity variance with use, stability of measurement and linearity of the current clinical standard Al2O3:C and a potential alternative, beryllium oxide. A set of ten optically stimulated luminescence dosimeters (OSLD), including five of each type, were used simultaneously and irradiated on a Versa HD linear accelerator. Having similar sensitivity, while Al2O3:C showed a relatively stable signal response from initial use, BeO was found to have a higher response to the same dose. However, BeO displayed a strong exponential decline from initial signal response following a model of $$Respons{e}_{BeO}=(0.55\pm 0.05){e}^{-\left(0.40\pm 0.05\right)x}+(0.54\pm 0.01)$$ R e s p o n s e BeO = ( 0.55 ± 0.05 ) e - 0.40 ± 0.05 x + ( 0.54 ± 0.01 ) , reaching stability after approximately 10 irradiation cycles. BeO was shown to have potentially higher accuracy than Al2O3:C, with less variation between individual doses. Both OSLD showed good linearity between 0.2–5.0 Gy. Between these bounds, Al2O3:C demonstrated a strong linear response following the trend $$Dose_{Al_{2}O_{3}, group(adj)}=(1.00\pm 0.09)x-(0.02\pm 0.04)\,{\text{Gy}}$$ D o s e A l 2 O 3 , g r o u p ( a d j ) = ( 1.00 ± 0.09 ) x - ( 0.02 ± 0.04 ) Gy , however beyond this showed deviation from linearity, resulting in a measured dose of $$12.0\pm 0.2$$ 12.0 ± 0.2 Gy at 10.0 Gy dose delivery. BeO showed strong linearity across the full examined range of 0.2–10.0 Gy with following a model of $$Dos{e}_{BeO, ind}=(0.98\pm 0.01)x+(0.04\pm 0.01)$$ D o s e B e O , i n d = ( 0.98 ± 0.01 ) x + ( 0.04 ± 0.01 ) Gy with a recorded dose at 10.0 Gy delivery as $$9.9\pm 0.1$$ 9.9 ± 0.1 Gy. In conclusion, BeO does show large variance in sensitivity between individual OSLD and a considerable initial variance and decline in dose–response, however after pre-conditioning and individual normalisation to offset OSLD specific sensitivity BeO provides not only a viable alternative to Al2O3:C, but potentially provide higher accuracy, precision and reproducibility for in-vivo dosimetry.

1 April 2023 in Journal of Medical Physics
Verified
Journal Article
Open access
PRPrabhakar Ramachandran
Prabhakar Ramachandran
TETamma Eswarlal
Tamma Eswarlal
MLMargot Lehman
Margot Lehman
ZCZachery Colbert
Zachery Colbert
Purpose: Optimizers are widely utilized across various domains to enhance desired outcomes by either maximizing or minimizing objective functions. In the context of deep learning, they help to minimize the loss function and improve model’s performance. This study aims to evaluate the accuracy of different optimizers employed for autosegmentation of non-small cell lung cancer (NSCLC) target volumes on thoracic computed tomography images utilized in oncology. Materials and Methods: The study utilized 112 patients, comprising 92 patients from “The Cancer Imaging Archive” (TCIA) and 20 of our local clinical patients, to evaluate the efficacy of various optimizers. The gross tumor volume was selected as the foreground mask for training and testing the models. Of the 92 TCIA patients, 57 were used for training and validation, and the remaining 35 for testing using nnU-Net. The performance of the final model was further evaluated on the 20 local clinical patient datasets. Six different optimizers, namely AdaDelta, AdaGrad, Adam, NAdam, RMSprop, and stochastic gradient descent (SGD), were investigated. To assess the agreement between the predicted volume and the ground truth, several metrics including Dice similarity coefficient (DSC), Jaccard index, sensitivity, precision, Hausdorff distance (HD), 95th percentile Hausdorff distance (HD95), and average symmetric surface distance (ASSD) were utilized. Results: The DSC values for AdaDelta, AdaGrad, Adam, NAdam, RMSprop, and SGD were 0.75, 0.84, 0.85, 0.84, 0.83, and 0.81, respectively, for the TCIA test data. However, when the model trained on TCIA datasets was applied to the clinical datasets, the DSC, HD, HD95, and ASSD metrics showed a statistically significant decrease in performance compared to the TCIA test datasets, indicating the presence of image and/or mask heterogeneity between the data sources. Conclusion: The choice of optimizer in deep learning is a critical factor that can significantly impact the performance of autosegmentation models. However, it is worth noting that the behavior of optimizers may vary when applied to new clinical datasets, which can lead to changes in models’ performance. Therefore, selecting the appropriate optimizer for a specific task is essential to ensure optimal performance and generalizability of the model to different datasets.

15 March 2022 in Biomedical Physics & Engineering Express
Verified
Journal Article
Closed access
RMRyan Motley
Ryan Motley
PRPrabhakar Ramachandran
Prabhakar Ramachandran
AFAndrew Fielding
Andrew Fielding
Purpose . The aim of this study was to assess the feasibility of the development and training of a deep learning object detection model for automating the assessment of fiducial marker migration and tracking of the prostate in radiotherapy patients. Methods and Materials . A fiducial marker detection model was trained on the YOLO v2 detection framework using approximately 20,000 pelvis kV projection images with fiducial markers labelled. The ability of the trained model to detect marker positions was validated by tracking the motion of markers in a respiratory phantom and comparing detection data with the expected displacement from a reference position. Marker migration was then assessed in 14 prostate radiotherapy patients using the detector for comparison with previously conducted studies. This was done by determining variations in intermarker distance between the first and subsequent fractions in each patient. Results . On completion of training, a detection model was developed that operated at a 96% detection efficacy and with a root mean square error of 0.3 pixels. By determining the displacement from a reference position in a respiratory phantom, experimentally and with the detector it was found that the detector was able to compute displacements with a mean accuracy of 97.8% when compared to the actual values. Interfraction marker migration was measured in 14 patients and the average and maximum ± standard deviation marker migration were found to be 2.0 ± 0.9 mm and 2.3 ± 0.9 mm , respectively. Conclusion . This study demonstrates the benefits of pairing deep learning object detection, and image-guided radiotherapy and how a workflow to automate the assessment of organ motion and seed migration during prostate radiotherapy can be developed. The high detection efficacy and low error make evident the advantages of using a pre-trained model to automate the assessment of the target volume positional variation and the migration of fiducial markers between fractions.

1 April 2021 in Journal of Medical Physics
Verified
Journal Article
Open access
PDF file
KAKeya Amarsee
Keya Amarsee
PRPrabhakar Ramachandran
Prabhakar Ramachandran
AFAndrew Fielding
Andrew Fielding
MLMargot Lehman
Margot Lehman
CNChristopher Noble
Christopher Noble
DNDaryl Ning
Daryl Ning
Purpose: Fiducial marker seeds are often used as a surrogate to identify and track the positioning of prostate volume in the treatment of prostate cancer. Tracking the movement of prostate seeds aids in minimizing the prescription dose spillage outside the target volume to reduce normal tissue complications. In this study, You Only Look Once (YOLO) v2TM (MathWorksTM) convolutional neural network was employed to train ground truth datasets and develop a program in MATLAB that can visualize and detect the seeds on projection images obtained from kilovoltage (kV) X-ray volume imaging (XVI) panel (ElektaTM). Methods: As a proof of concept, a wax phantom containing three gold marker seeds was imaged, and kV XVI seed images were labeled and used as ground truth to train the model. The projection images were corrected for any panel shift using flex map data. Upon successful testing, labeled marker seeds and projection images of three patients were used to train a model to detect fiducial marker seeds. A software program was developed to display the projection images in real-time and predict the seeds using YOLO v2 and determine the centers of the marker seeds on each image. Results: The fiducial marker seeds were successfully detected in 98% of images from all gantry angles; the variation in the position of the seed center was within ± 1 mm. The percentage difference between the ground truth and the detected seeds was within 3%. Conclusion: Our study shows that deep learning can be used to detect fiducial marker seeds in kV images in real time. This is an ongoing study, and work is underway to extend it to other sites for tracking moving structures with minimal effort.

4 January 2024 in Biomedical Physics & Engineering Express
Verified
Journal Article
Open access
PDF file
ZCZachery Morton Colbert
Zachery Morton Colbert
DADaniel Arrington
Daniel Arrington
MFMatthew Foote
Matthew Foote
JGJonas Gårding
Jonas Gårding
DFDominik Fay
Dominik Fay
PRPrabhakar Ramachandran
Prabhakar Ramachandran
Objective: Automated medical image segmentation (MIS) using deep learning has traditionally relied on models built and trained from scratch, or at least fine-tuned on a target dataset. The Segment Anything Model (SAM) by Meta challenges this paradigm by providing zero-shot generalisation capabilities. This study aims to develop and compare methods for refining traditional U-Net segmentations by repurposing them for automated SAM prompting. Approach: A 2D U-Net with EfficientNet-B4 encoder was trained using 4-fold cross-validation on an in-house brain metastases dataset. Segmentation predictions from each validation set were used for automatic sparse prompt generation via a bounding box prompting method (BBPM) and novel implementations of the point prompting method (PPM). The PPMs frequently produced poor slice predictions (PSPs) that required identification and substitution. A slice was identified as a PSP if it (1) contained multiple predicted regions per lesion or (2) possessed outlier foreground pixel counts relative to the patient’s other slices. Each PSP was substituted with a corresponding initial U-Net or SAM BBPM prediction. The patients’ mean volumetric dice similarity coefficient (DSC) was used to evaluate and compare the methods’ performances. Main results: Relative to the initial U-Net segmentations, the BBPM improved mean patient DSC by 3.93 ± 1.48% to 0.847 ± 0.008 DSC. PSPs constituted 20.01–21.63% of PPMs’ predictions and without substitution performance dropped by 82.94 ± 3.17% to 0.139 ± 0.023 DSC. Pairing the two PSP identification techniques yielded a sensitivity to PSPs of 92.95 ± 1.20%. By combining this approach with BBPM prediction substitution, the PPMs achieved segmentation accuracies on par with the BBPM, improving mean patient DSC by up to 4.17 ± 1.40% and reaching 0.849 ± 0.007 DSC. Significance: The proposed PSP identification and substitution techniques bridge the gap between PPM and BBPM performance for MIS. Additionally, the uniformity observed in our experiments’ results demonstrates the robustness of SAM to variations in prompting style. These findings can assist in the design of both automatically and manually prompted pipelines.

15 March 2021 in Biomedical Physics & Engineering Express
Verified
Journal Article
Closed access
PDF file
PRPrabhakar Ramachandran
Prabhakar Ramachandran
CNChristopher Noble
Christopher Noble
CLChristian Langton
Christian Langton
BPBen Perrett
Ben Perrett
JCJoshua Cox
Joshua Cox
GMGeorge McGill
George McGill
Geometric distortions in magnetic resonance can introduce significant uncertainties into applications such as radiotherapy treatment planning and need to be assessed as part of a comprehensive quality assurance program. We report the design, fabrication, and imaging of a custom 3D printed unibody MR distortion phantom along with quantitative image analysis. Methods : The internal cavity of the phantom is an orthogonal three-dimensional planar lattice, composed of 3 mm diameter rods spaced equidistantly at a 20 mm centre-centre offset repeating along the X, Y, and Z axes. The phantom featured an overall length of 308.5 mm, a width of 246 mm, and a height of 264 mm with lines on the external surface for phantom positioning matched to external lasers. The MR phantom was 3D printed in Nylon-12 using an advancement on traditional selective laser sintering (SLS) (HP Jet Fusion 3D—4200 machine). The phantom was scanned on a Toshiba Aquilion CT scanner to check the integrity of the 3D print and correct for any resultant issues. The phantom was then filled with NiSO 4 solution and scanned on a 3T PET-MR Siemens scanner for selected T1 and T2 sequences, from which distortion vectors were generated and analysed using in-house software written in Python. Results : All deviations of the node positions from the print design were less than 1 mm, with an average displacement of 0.228 mm. The majority of the deviations were smaller than the 0.692 mm pixel size for this dataset. Conclusion : A customised 3D printed MRI-phantom was successfully printed and tested for assessing geometric distortion on MRI scanners. 3D printed phantoms can be considered for clinics wishing to assess geometric distortions under specific conditions, but require resources for design, fabrication, commissioning, and verification.