Mustafa Sami, Ph.D.

Email: m.sami(at)oist.jp

Professional Experience

  • RIKEN Center for Biosystems Dynamics Research, Kobe (BDR)
  • RIKEN Center for Developmental Biology, Kobe (CDB)
  • Nippon Boehringer Ingelheim Co., LTD.
  • Tampere University of Technology
  • Niigata University

Research Interest

Signal Processing, Digital Image Processing, Complex Network Analysis, Deep Learning, Machine Learning, Pattern Recognition, Quantitative Biology, Computer-Aided Diagnosis, Image Segmentation, Feature Selection and Extraction, Video and Text Processing and Retrieval.

Teaching Experience

  • Digital Image Processing for Quantitative Biology, RIKEN Kobe
  • Image Processing Using MATLAB, RIKEN Kobe
  • Using AI for Clinical Applications, Kobe Institute of Computing (KIC)

Research Activities

Cell and Developmental Biology

- Leading the development of an image-based software systems and GUI applications to process and analyse the confocal microscopic images. Here we tried to understand the early cell shape changes during organ growth and development in Drosophila. These images are very noisy and hard to extract important features from them. A specified intensity and morphological filters were designed to help biologists to observe and track cells in a fully automatic manner.

  • Mustafa M. Sami, Yosuke Ogura, Yu-Chiun Wang, and Shigeo Hayashi, “Automated image processing and analysis software for epithelial cells quantification” 17th International European Light Microscopy Initiative Meeting (ELMI 2017), Dubrovnik, Croatia, May 2017.
  • Michiko Takeda, Mustafa M. Sami, Yu-Chiun Wang, A homeostatic apical microtubule network shortens cells for epithelial folding via a basal polarity shift, Nature Cell Biology, Vol.20, pp. 36-45, Dec.2017.
  • Yosuke Ogura, Fu-Lai Wen, Mustafa M. Sami, Tatsuo Shibata, Shigeo Hayashi, A Switch-like Activation Relay of EGFR-ERK Signaling Regulates a Wave of Cellular Contractility for Epithelial Invagination. Developmental Cell, Vol.46, pp. 162-172, July 2018.
  • Anthony S Eritano, Claire L Bromley, Antonio Bolea Albero, Lucas Schütz, Fu-Lai Wen, Michiko Takeda, Takashi Fukaya, Mustafa M. Sami, Tatsuo Shibata, Steffen Lemke, and Yu-Chiun Wang. Tissue-scale mechanical coupling reduces morphogenetic noise to ensure precision during epithelial folding. Developmental Cell, Vol.53, pp. 212–228, April 2020.

- Leading the development of automated computer software system to quantify the signals of fluorescence energy transfer (FRET) reporters for activated extracellular signal-regulated kinase (ERK), and used this system to monitor the spatio-temporal dynamics of ERK during neuroectoderm patterning in Drosophila embryos.

  • Yosuke Ogura#, Mustafa M. Sami#, Housei Wada, Shigeo Hayashi, Automated FRET quantification reveals distinct subcellular ERK activation kinetics in response to graded EGFR signaling in Drosophila. Genes to Cells, Vol.24, pp. 297–306, Feb. 2019. (# contributed equally)

- Leading the image processing and analysis of Electron Microscopy (EM) images for the nanopore formation project applied to the cuticle of the insect olfactory sensillum system.

  • Toshiya Ando, Sayaka Sekine, Sachi Inagaki, Kazuyo Misaki, Laurent Badel, Hiroyuki Moriya, Mustafa M. Sami, Yuki Itakura, Takahiro Chihara, Hokto Kazama, Shigenobu Yonemura, Shigeo Hayashi. Nanopore Formation in the cuticle of an insect olfactory sensillum. Current Biology, Vol.29, pp. 1–9, May 2019.

- Leading the development of a quantification method applied to super-resolution live imaging of super-cellular circumferential actin cable formation during tracheal tubulogenesis.

  • Sayaka Sekine, Mustafa M. Sami, Housei Wada, Shigeo Hayashi, “Super-resolution live imaging of supercellular circumferential actin cable formation during tracheal tubulogenesis”, Joint Annual Meeting of JSDB and JSCB, Tokyo, Jun. 2018.
  • Sayaka Sekine, Mitsusuke Tarama, Housei Wada, Mustafa M. Sami, Tatsuo Shibata & Shigeo Hayashi,  "Emergence of periodic circumferential actin cables from the anisotropic fusion of actin nanoclusters during tubulogenesis", Nature Communications, Jan. 2024.

Heart-on-a-chip Micro Device

- Leading the development of a software tool to track nano-bead particles for a new device that generates heart tissues from human iPS cells, and can be used for drug discovery for heart diseases and cardiac toxicity tests. These nano-beads are very small and hard to be tracked because of their irregular speed movements in a difficult tissue environments.

  • Mosha Abulaiti, Yaxiaer Yalikun, Kozue Murata, Asako Sato, Mustafa M. Sami, Yuko Sasaki, Yasue Fujiwara, Kenji Minatoya, Yuji Shiba, Yo Tanaka & Hidetoshi Masumoto. Establishment of a heart-on-a-chip microdevice based on human iPS cells for the evaluation of human heart tissue function. Scientific Reports, 10, 19201, Nov. 2020.

Clinical Pathology  

- Leading the development of a new computer-aided diagnostic (CAD) system for the automatic grading of borderline grades of human oral cancer. Epithelial dysplasia and carcinoma in-situ of the oral mucosa are two different borderline grades similar to each other, and it is difficult for pathologists to distinguish these two lesions in hematoxylin and eosin-stained sections only. To support objective differential diagnoses, we developed this new image-based CAD system.

  • Mustafa M. Sami, Masahisa Saito, Shogo Muramatsu, Hisakazu Kikuchi, and Takashi Saku, “A Computer-Aided Diagnostic Method of Borderline Grades of Oral Cancer,” IEICE Transactions on Fundamentals. Vol. E93-A, No.8, pp. 1544-1552, Aug. 2010.
  • Mustafa M. Sami, Masahisa Saito, Shogo Muramatsu, Toshihiko Mikami, Kamal Al-Eryani, Jun Cheng, Faleh A. Sawair, Rasha Abu Eid, Hisakazu Kikuchi, and Takashi Saku, “Twin-pair rete ridge analysis: a computer-aided method for facilitating objective histopathological distinction between epithelial dysplasia and carcinoma in-situ of the oral mucosa,” Oral Medicine & Pathology, Vol.14, pp.89-98, Jan. 2010.
  • Mustafa M. Sami, M. Saito, H. Kikuchi, and T. Saku, “A Computer-Aided Distinction of Borderline Grades of Oral Cancer,” Presented in IEEE International Conference on Image Processing (ICIP 2009), Cairo, Egypt, Nov. 2009.
  • Mustafa M. Sami, M. Saito, H. Kikuchi, and T. Saku, “Twin Rete Ridge Analysis for Histopathological Diagnosis in H&E Stained Microscopic Images of the Oral Mucosa,” Presented in 6th International Workshop on Computational Systems Biology, (WCSB’09), Arhus, Denmark, Jun. 2009.
  • Mustafa M. Sami, H. Kikuchi, and T. Saku, “Shape Analysis of the Rete Processes for Borderline Malignancies of the Oral Mucosal Epithelia,” Presented in 8th International Symposium on Computer Methods in Biomechanics and Biomedical Engineering, (CMBBE’08) Porto, Portugal, Mar. 2008.

Drug Discovery    

- Leading the development of a new CAD system at Boehringer Ingelheim Pharmaceutical Company for a new drug target validation method applied to “lung fibrosis” in human. We used M&T stain microscopic images for the development of this system based on Ashcroft standardizing technique. (Unpublished work owned by Boehringer Ingelheim).

Brain Science

- Leading the development of image-processing and analysis tool to quantify the pathophysiological changes in fatigued rat. Our aim was to identify the neurons involve in fatigue loading and recovery by examining the c-Fos expression levels in histological sections microscopic images of the brain tissues.

  • Mustafa M. Sami, Ko-hei Akazawa, Yilong Cui, Hisakazu Kikuchi, and Yosky Kataoka, Multi-Atlas Applications in Fatigue Pathophysiology, 15th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2012), Nice, France, Oct. 2012.
  • Mustafa M. Sami, Yasuhisa Tamura, Yilong Cui, Hisakazu Kikuchi, and Yosky Kataoka, “3D Quantification of c-Fos Immunopositive Cells for Fatigue Screening, “CDB Symposium 2012 on Quantitative Development Biology, Kobe, Japan, Mar. 2012.

- Leading the development of a new automatic quantification method to automatically count the number of targeted fluorescently labeled molecules of NG2-glial cells observed from confocal microscopic images of rats. NG2-glial cells were monitored using our system in order to identify the type of proliferation that may produce another cells of the same type or to a different type of cells known as astrocyte cells.

  • Mustafa M. Sami, Yasuhisa Tamura, Hisakazu Kikuchi, and Yosky Kataoka, “In-vitro Cell Quantification Method Based on Depth Dependent Analysis of Brain Tissue Microscopic Images, “33rd Annual International Conference of IEEE Engineering in Medicine and Biology Society (EMBC 2011), Boston, USA, Aug. 2011.
  • Tamura Yasuhisa, Eguchi Asami, Jin Guanghua, Mustafa M. Sami, Kataoka Yosky, Cortical spreading depression shifts cell fate determination of progenitor cells in the adult cortex, Journal of Cerebral Blood Flow & Metabolism, 32(10):1879-87, Oct. 2012.

Video and Text Retrieving   

- Co-leading the development of a text retrieving system for prior art search in the USPTO patent database.

  • John Parker, Mustafa M. Sami, Jonathan Miller, Mei Kobayashi, “Patent Classification using Balanced Input for Naive Bayes”, The 74th Joint Conference of Electrical, Electronics and Information Engineers in Kyushu International Session, Sojo University, Kumamoto City, Japan, September 8, 2023.

- Co-leading the development of a video retrieving system from video databases using a new hierarchical clustering analysis method of feature vectors that are derived from wavelet coefficients of video frames. Here we constructed a storyboard that contains a user-specified number of key-frames from a given Motion-JPEG 2000 sequence.

  • Satoshi Hasebe, Mustafa M. Sami, Shogo Muramatsu, Hisakazu Kikuchi, “Constructing storyboards based on hierarchical clustering analysis,” Proc. SPIE, The International Society for Optical Engineering, Vol. 5960, pp. 437-445, Jan. 2010.
  • Satoshi Hasebe, Mustafa M. Sami, Shogo Muramatsu, and Hisakazu Kikuchi, “Constructing Storyboards Based on Hierarchical Clustering Analysis,” Presented at Visual Communications and Image Processing (VCIP 2005), Beijing, China, Jul. 2005.

Deep Learning  

- Leading the development of a new deep learning based classification method to identify the morphologenetic behaviors from 3D epithelial cell shape changes. We used support vector machine (SVM) to optimize ResNet-101 deep neural network in order to identify 3 groups of epithelial cells in a fully automatic way.

  • Mustafa M. Sami and Shigeo Hayashi, “AI classifies epithelial cells into distinct groups of morphogenetic behaviors” (Under revision).
  • Mustafa M. Sami, Takuya Maeda, Shigeo Hayashi, “3D cell shape recognition using AI”, Joint Annual Meeting of JSDB and JSCB, Tokyo, Jun. 2018.