Shenghua He

å adjflf@r.postjobfree.com D @shenghua-he H @shenghh2015 @Shenghua He

Education

Washington Univesrity in St. Louis St. Louis, MO

PhD. Computer Science Aug. 2016 – 2021

• GPA, 3.92/4.0 Advisors: Mark Anastasio (UIUC), Hua Li (UIUC) University of Illinois at Urbana-Champaign Urbana, IL PhD. Visiting Student Apr. 2019 – 2021

• Advisors: Mark Anastasio, Hua Li

Beijing University of Posts and Telecommunications Beijing, China M.E. Electronic Communication and Engineering Sep. 2012 – Apr. 2015

• Advisor: Xiangming Wen

Wuhan University of Technology Wuhan, China

B.E. Electronic Science and Engineering Aug. 2008 – Jun. 2012

• GPA: 90.4/100, Ranking: top 3.7%

Expertise

Deep learning

• Extensive practices on building convolutional neural network (CNN) models for computer vision tasks:

– EfficientNets, ResNet, VGG, AlexNet for image classification

– U-net, FCN, Attention U-net, EfficientNet U-net for image segmentation

– Mask R-CNN, R-CNN, and YOLO for object detection

– Fully regression neural networks (FCRN) for object counting

• Extensive practices on CNN model training: deep supervision, transfer learning, adversarial learning, dropout

• Good practices on generative adversarial networks (GAN): progressive GAN, cycleGAN

• Other good practices: Autoencoder and proGAN for image anomaly detection

• Familiar with recurrent neural networks (RNN) for building up sequential models: LSTM for natural language processing

Machine Learning

• Supervised algorithms: Linear regression, Logistic regression, SVM, KNN, Evidential-KNN

• Unsupervised algorithms: K-mean, PCA, Canonical correlation analysis (CCA) Biomedical image analysis (deep learning application)

• Processed 2D/3D experimental biomedical images that contain noise, low contrast, reconstruction artifacts and tissue complex background

• Addressed/mitigated challenges that include small-size dataset, unbalanced classes, ambiguous inter-class objects in training deep models on biomedical images Unimodal/multimodal cancer treatment outcome prediction

• Selected sparse set of informative miRNA features from small-size dataset with large amounts of features to learn models for robust cancer treatment outcome prediction

• Fused correlated information from multimodal data (image modality and miRNA features) to learn models for classifying whether cancer treatment is likely to succeed Engineering skills

• Languages: Python: Numpy, Keras, Tensorflow, Pytorch; Matlab; Java

• Tools: Linux; HPC; Docker; Git; Jupyter Notebooks Experience and projects

University of Illinois at Urbana-Champaign Apr. 2019 – Present PhD Visiting Student Urbana, IL

• Proposed and developed a novel CNN model (CNN/U-net/EfficientNets) for multi-class biomedical image segmentation.

• Proposed and developed novel CNN model (CNN/U-net/EfficientNets) for biomedical image modality translation.

• Proposed and developed novel unimodal/multimodal learning models for cancer treatment outcome prediction.

Washington University in St. Louis Aug. 2016 – present Research Assistant St. Louis, MO

• Proposed and developed a novel deep novel deep learning model (CNN/U-net/FCRN) to estimate cell density map for counting high-density and occluded cells in biomedical images.

• Developed a CNN model for semantic medical image segmentation. Clemson University May 2015 – May 2016

Research Assistant Clemson, SC

• Proposed and developed a traffic redundancy elimination system for wireless networks. Beijing University of Posts and Telecommunications Sep. 2012 – Apr. 2015 Research Assistant Beijing, China

• Proposed and develop wireless resource allocation algorithms for heterogeneous wireless networks.

Publications

Journal

• Maliazurina Saad*, Shenghua He*, Wade Thorstad, Hiram Gay, Su Ruan, Xiaowei Wang, and Hua Li, ”Learning-based Cancer Treatment Outcome Prognosis using Multimodal Biomarkers”,

(submitted to IEEE Transactions on Radiation and Plasma Medical Sciences 2021)(* equal contribution)

• Shenghua He, Chunfeng Lian, Wade Thorstad, Hiram Gay, Yujie Zhao, Su Ruan, Xiaowei Wang, and Hua Li,”A novel machine learning approach for cancer treatment prognosis and its applications in oropharyngeal cancer with microRNA biomarkers.”(Bioinformatics 2021 under review)

• Chenfei Hu*, Shenghua He*, Young Jae Lee, Yuchen He, Edward Minjae Kong, Hua Li, Mark A. Anastasio, and Gabriel Popescu. ”Label-free cell viability assay using phase imaging with computational specificity.” bioRxiv (2020). (Nature communications 2021 under review)

• Kyaw Thu Minn, Yuheng C. Fu, Shenghua He, Sabine Dietmann, Steven C. George, Mark A. Anastasio, Samantha A. Morris, and Lilianna Solnica-Krezel. ”High-resolution transcriptional and morphogenetic profiling of cells from micropatterned human ESC gastruloid cultures.” Elife 9 (2020): e59445.

• Shenghua He, Kyaw Thu Minn, Lilianna Solnica-Krezel, Mark A. Anastasio, and Hua Li.

”Deeply-supervised density regression for automatic cell counting in microscopy images.” Medical Image Analysis 68 (2020): 101892.

• Shenghua He*, Jian Wu*, Chunfeng Lian, H. Michael Gach, Sasa Mutic, Walter Bosch, Jeff Michalski, and Hua Li. ”An Adaptive Low-Rank Modeling-based Active Learning Method for Medical Image Annotation.” IRBM (2020).

• Yan Zhuang, Lei Yu, Haiying Shen, William Kolodzey, Nematollah Iri, Gregori Caulfield, and Shenghua He. ”Data collection with accuracy-aware congestion control in sensor networks.” IEEE Transactions on Mobile Computing 18, no. 5 (2018): 1068-1082.

• Jun Zhao, Zhaoming Lu, Xiangming Wen, Haijun Zhang, Shenghua He, and Wenpeng Jing.

”Resource management based on security satisfaction ratio with fairness-aware in two-way relay networks.” International Journal of Distributed Sensor Networks 11, no. 7 (2015): 819195.

• Shenghua He, Zhaoming Lu, Xiangming Wen, Zhicai Zhang, Jun Zhao, and Wenpeng Jing. ”A pricing power control scheme with statistical delay QoS provisioning in uplink of two-tier OFDMA femtocell networks.” Mobile Networks and Applications 20, no. 4 (2015): 413-423. Conference papers

• Chenfei Hu*, Shenghua He*, Young Jae Lee, Yuchen R. He, Mark Anastasio, and Gabriel Popescu, ”Label-free cell viability assay using phase imaging with computational specificity

(PICS)”, SPIE BiOS 2021.

• Yuchen R. He*, Shenghua He*, Mikhail E. Kandel*, Young Jae Lee, Nahil Sobh, Mark Anastasio, and Gabriel Popescu, ”Cell cycle detection using phase imaging with computational specificity

(PICS)”, SPIE BiOS 2021.

• Zong Fan, Shenghua He, Su Ruan, Xiaowei Wang, and Hua Li, ”Deep learning-based multi-class COVID-19 classification with x-ray Images”, SPIE Medical Imaing 2021.

• Fu Li, Umberto Villa, Seonyeong Park, Shenghua He, Mark A. Anastasio, ”A framework for ultrasound computed tomography virtual imaging trials that employs anatomically realistic numerical breast phantoms”, SPIE Medical Imaing 2021.

• Shenghua He, Weimin Zhou, Hua Li, and Mark A. Anastasio. ”Learning numerical observers using unsupervised domain adaptation,” SPIE Medical Imaging 2020.

• Shenghua He, Kyaw Thu Minn, Lilianna Solnica-Krezel, Hua Li, and Mark Anastasio.

”Automatic microscopic cell counting by use of unsupervised adversarial domain adaptation and supervised density regression”,SPIE Medical Imaging 2019.

• Shenghua He, Ziyao Yi, Su Ruan, Mark Anastasio, Sasa Mutic, Hiram Gay, Wade Thorstad, Xiaowei Wang, Hua Li,”MicroRNA-Based Survival and Relapse Prognosis for Oropharyngeal Cancer Treatment by Use of Cox Regression and Belief Function Theory”, AAPM 2019

• Shenghua He, Kyaw Thu Minn, Lilianna Solnica-Krezel, Mark Anastasio, and Hua Li.

”Automatic microscopic cell counting by use of deeply-supervised density regression model,” SPIE Medical Imaging 2019.

• Shenghua He, Jie Zheng, Akiko Maehara, Gary Mintz, Dalin Tang, Mark Anastasio, and Hua Li.

”Convolutional neural network based automatic plaque characterization for intracoronary optical coherence tomography images”, SPIE Medical Imaging 2018.

• Shenghua He, Haiying Shen, Vivekgautham Soundararaj, and Lei Yu. ”Cloud Assisted Traffic Redundancy Elimination for Power Efficiency in Smartphones.” In 2018 IEEE 15th International Conference on Mobile Ad Hoc and Sensor Systems (MASS), pp. 371-379. IEEE, 2018.

• Shenghua He, Ling Zhang, Xiangming Wen, Zhicai Zhang, Zhaoming Lu, and Yong Sun.

”Price-based power control with statistical delay QoS guarantee in two-tier femtocell networks.” In 2014 21st International Conference on Telecommunications (ICT), pp. 318-322. IEEE, 2014.

Contact this candidate