I have found some of my most rewarding work through my reserach, where I was allowed to pursue projects that fully utilized knowledge from all three of my degrees. These wonderful opportunities have allowed me to be published multiple times as a primary author, present on a podium in a national conference, and implement groundbreaking analysis within a new discipline. Below I have information from both the labs I have worked in as well as the papers that I have directly wrote/ contributed to.
Computational Ecology, Evolution and Biology Lab - University of Michigan
Published open-source R package titled weightedClustSuite [R, C++, Python] for Machine Learning Density Clustering & Validation, used on weighted species abundance data enabling identification of species distribution clusters Link
Encoded Density Clustering ML Model, selected with DBCV analysis suite to identify coexistence among competing species contrary to existing principles of competitive exclusion. Model architecture and validation designed from recent ML research.
Operate as an interdisciplinary reference within lab to explain and analyze machine learning and data analysis research.
Findings resulted as Primary Author in two Publications, was selected for Poster, Oral Presentation in SPIE Conference 2019
Performed Feature Extraction and Selection on Bladder CTU Scan Segmentations to extract key features in Bladder Cancer Lesion Staging (staging determines the avenue of treatment)
Implemented and trained a series of Back Propagated Neural Networks, Linear Discriminant Analysis, Support Vector Machines, and Random Forest Algorithms – validated through ROC analysis - resulting in a models’ cancer staging accuracy (+80%)
Notable Technologies Used: C++, C, Python, Weka, MS Excel
I am currently working on a paper in which I am developing an R package for Unsupervised ML Biological Trait Density Clustering using a recent clustering algorithm from 2018 research and a Density Based Cross Validatiion metric from 2016 research
My research/code was ustilized as a secondary author in the implementation of a Neural Network in feature detecion of bladder cancer staging
Recommended citation: Daniel Hoklai Chapman-Sung, Lubomir Hadjiiski, Dhanuj Gandikota, Heang-Ping Chan, Ravi Samala, Elaine M. Caoili, Richard H. Cohan, Alon Weizer, Ajjai Alva, and Chuan Zhou "Convolutional neural network-based decision support system for bladder cancer staging in CT urography: decision threshold estimation and validation", Proc. SPIE 11314, Medical Imaging 2020: Computer-Aided Diagnosis, 113141T (16 March 2020); https://doi.org/10.1117/12.2551309
Primary Author - My research found that the use of decision threholds with multiple ML classifiers [LDA, BPNN, RAF, SVM] in our diagnosis system accurately diagnosed bladder cancer stages
Recommended citation: Dhanuj Gandikota, Lubomir Hadjiiski, Heang-Ping Chan, Kenny H. Cha, Ravi Samala, Elaine M. Caoili, Richard H. Cohan, Alon Weizer, Ajjai Alva, Chintana Paramagul, Jun Wei, and Chuan Zhou "Bladder cancer staging in CT urography: estimation and validation of decision thresholds for a radiomics-based decision support system", Proc. SPIE 10950, Medical Imaging 2019: Computer-Aided Diagnosis, 109500W (13 March 2019); https://doi.org/10.1117/12.2513566
Primary Author - My research found that ML linear discriminant analysis was statistiscally significant in sorting bladder cancer staging determining different treatment avenues.
Recommended citation: Dhanuj Gandikota, Lubomir Hadjiiski, Kenny H. Cha, Heang-Ping Chan, Elaine M. Caoili, Richard H. Cohan, Alon Weizer, Ajjai Alva, Chintana Paramagul, Jun Wei, and Chuan Zhou "Bladder cancer staging in CT urography: effect of stage labels on statistical modeling of a decision support system", Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis, 105752B (27 February 2018); https://doi.org/10.1117/12.2295013
