I, Nitin Satpute, am an Architect Specialist AI Infrastructure GPU. I have done PhD in Computer Science from University of Cordoba, Spain under Marie-Curie ESR Fellowship funded by EU. I have completed Master of Engineering (ME) in Embedded Systems from Birla Institute of Technology and Science (BITS), Pilani in June 2013. I have worked at Indian Institute of Science (IISc) Bangalore, University of Siena Italy, NTNU and Oslo University Hospital Norway and Aarhus University Denmark. My research interests include HPC, GPU programming using CUDA, AI, Deep Learning, Machine Learning and Analog Computing aimed for Spiking Neural Networks.
I have attended a) MIT GSW at Novotel Hyderabad Convention Centre (Mar, 2016) b) Deep Learning Training program presented by Deep Learning Institute, nVIDIA & hosted by GPU Center of Excellence, IIT Bombay on Dec 05, 2016 c) “The 2015 LOFAR Surveys Meeting” held at Leiden, Netherlands (Sept, 2015) d) “AXIOM Face to Face Meet” held at Barcelona Super-computing Center (BSC), Spain (Jun, 2015). I have assisted Prof. Donald Reay from Heriot-Watt University, UK in conducting Faculty Development Program on DSP for Educators at IIIT Bangalore, VNIT Nagpur and NIT Patna (Mar & Sep, 2016).
I have been involved in the development and implementation of fast parallel algorithms for liver enhancement and segmentation on GPUs. I have worked on GPU, Deep Learning, Medical Imaging and implemented UNet for liver segmentation, YOLO for object detection and GAN based Pix2pix for image translation on GPU(s). I have following publications in High Performance Medical Image Processing using environments and platforms such as Linux, C++, CUDA, OpenCL, Python, and NVIDIA GPU GeForce GTX 1050, RTX 2070 and 2080 Ti.
Publication on Image Enhancement and Segmentation:
1. Nitin Satpute, R. Naseem, E. Pelanis, J. Gomez-Luna, F. Alaya Cheikh, O. J. Elle, and J. Olivares, “GPU acceleration of liver enhancement for tumor segmentation”, published in Computer Methods and Programs in Biomedicine 184 (2020) 105285, DOI: https://doi.org/10.1016/j.cmpb.2019.105285. (CMPB – Q1, IF 3.424).
Publication on Image Segmentation:
2. Nitin Satpute, R. Naseem, R. Palomar, O. Zachariadis, J. Gomez-Luna, F. Alaya Cheikh, and J. Olivares, “Fast Parallel Vessel Segmentation” published in Computer Methods and Programs in Biomedicine (2020) 105430, DOI: https://doi.org/10.1016/j.cmpb.2020.105430. (CMPB – Q1, IF 3.424).
Publication on Image Registration:
3. O. Zachariadis, A. Teatini, Nitin Satpute, J. Gomez-Luna, O. Mutlu, O. J. Elle, and J. Olivares, “Accelerating B-spline Interpolation on GPUs: Application to Medical Image Registration”, published in Computer Methods and Programs in Biomedicine (2020) 105431, DOI: https://doi.org/10.1016/j.cmpb.2020.105431. (CMPB – Q1, IF 3.424).
Earlier, I have worked on Convolutional Deep Neural Network (CNN) for breast cancer detection at at VNIT, Nagpur. I have implemented convolutional, activation (ReLU and Sigmoid), pooling, fully connected and softmax layers. I have developed mathematical models for robust feature learning. It includes learning from adversarial examples and making classifiers perform robustly when confronted with such hard positive/negative examples. The parameters responsible for feature learning are number of each of the mentioned layers, selection of the activation functions, pooling parameters (max or mean, stride depth, pool size) and the type and size of the filters. The flow starts by choosing number of parameters randomly and initializing them with random numbers. Then the experimental and mathematical analysis is done to update the filters. I have implemented back-propagation algorithm (stochastic gradient descent (SGD)) for error minimization. (Environment – Linux, Python and MATLAB)
During my research tenure at University of Siena, I have developed a sound foundation of CUDA parallel programming model in the project GPU based implementation of Matrix Multiplication (MM). I have presented posters in PUMPS 2015, Barcelona Supercomputing Center (BSC), Spain and ACACES 2015, Fiuggi, Italy. I have published a research paper on “Comparing and Evaluating GPU Platforms, using a Single Point” in ACACES 2015.
Having worked as a project associate at Supercomputer Education and Research Centre (SERC), Indian Institute of Science (IISc) Bangalore, India, I have published a research paper on the results obtained from the project “Flexible Scalable hardware architecture for radial basis function neural networks (RBFNN)” in the International Conference on VLSI Design and Embedded Systems, 2015. I have worked on the Bulk Synchronous Parallel (BSP) Computation Model and the implementation of H.264 & H.265 video decoder. The implementation of H.264 video decoder supports baseline profile capable of decoding HD video.
As a post-graduate student at BITS Pilani, India, I have studied probability, expectations and distributions (Joint, Gaussian, Bayes) in the course Advanced Digital Signal Processing (ADSP). I have published a paper on “Tabu Search based implementation of object tracking using Joint Colour Texture Histogram” in the International Conference on Industrial and Information Systems, IIT Chennai, IEEE Explore.
During the course of Bachelor of Engineering (BE) program from the Yeshwantrao Chavan College of Engineering (YCCE), Nagpur, India, I have gained valuable experience in Image Processing and Engineering Mathematics. I have studied Probability, Random Variables, Distributions, and Transformations (Laplace, Fourier etc) in Engineering Mathematics. I have explored RBFNN, Bayesian Clustering and Simulated Annealing in the project “Segmentation & Classification of MRI Brain Images Using Texture Features” and published a research paper in the International Journal of Machine Intelligence and Applications.
Apart from these works, I am also a tutor for IIT-JEE and GATE aspirants. I provide guidance for IIT-JEE Mathematics and Physics. I have taught wide range of topics including Linear Algebra, Calculus, Probability, Coordinate Geometry, Atoms Molecules & Nuclei, Gravitation, Current Electricity, Magnetism, Electromagnetic Induction, Optics etc. The Indian Institutes of Technology Joint Entrance Examination (IIT-JEE) is an engineering entrance examination in India, conducted annually.
IIT-JEE is recognized as one of the toughest examinations in the world as it has very low admission rate. (link: https://en.wikipedia.org/wiki/Indian_Institute_of_Technology_Joint_Entrance_Examination).
GATE is Graduate Aptitude Test in Engineering for admissions to various post graduate programs in Indian higher education institutes. GATE scores are also being used by public sector units and government agencies for recruiting graduates in entry level positions (link: https://en.wikipedia.org/wiki/Graduate_Aptitude_Test_in_Engineering).
Nitin R Satpute 