TY - GEN
T1 - Supervised Regression Study for Electron Microscopy Data
AU - Al-Haija, Qasem Abu
AU - Nasr, Kamal Al
N1 - Funding Information:
-----------------------------------------------------------------------------------------* Corresponding Author. This work was supported by the US National Science Foundation (NSF) Research Initiation Award (RIA) (HRD: 1600919) and the NIH Research grant (R15-AREA: 1R15GM126509-01).
Publisher Copyright:
© 2019 IEEE.
PY - 2019/11
Y1 - 2019/11
N2 - This study presents a supervised regression model to estimate the growth of Electron Microscopy experimental data for a decade ahead. The study employs the autoregression process model using the best curve-fitting that optimizes the level of confidence. Further, the proposed model retains the smallest normalized estimation error. The developed model was competently utilized to estimate the size of Electron Microscopy (EM) data expected to be released within years 2019-2028. One EM dataset was used to model and predict the annual growth of released 3DEM. Another EM dataset was used to model and predict the annual number of 3-Dimensional EM achieving resolution 10 A or better. Indeed, both models used EM data collected in the past 18 years, 2002-2018. The experimental results showed that the best curve-fitting orders to predict both datasets were AR(5) at 96.8% and AR(6) at 85% for the released 3DEM and 3DEM resolutions datasets, respectively. Therefore, the estimation findings disclose an exponential growing performance in the upcoming evolution for both, the released 3DEM and 3DEM resolutions datasets. However, the evolution rate of the released 3DEM confirms a faster exponential growth.
AB - This study presents a supervised regression model to estimate the growth of Electron Microscopy experimental data for a decade ahead. The study employs the autoregression process model using the best curve-fitting that optimizes the level of confidence. Further, the proposed model retains the smallest normalized estimation error. The developed model was competently utilized to estimate the size of Electron Microscopy (EM) data expected to be released within years 2019-2028. One EM dataset was used to model and predict the annual growth of released 3DEM. Another EM dataset was used to model and predict the annual number of 3-Dimensional EM achieving resolution 10 A or better. Indeed, both models used EM data collected in the past 18 years, 2002-2018. The experimental results showed that the best curve-fitting orders to predict both datasets were AR(5) at 96.8% and AR(6) at 85% for the released 3DEM and 3DEM resolutions datasets, respectively. Therefore, the estimation findings disclose an exponential growing performance in the upcoming evolution for both, the released 3DEM and 3DEM resolutions datasets. However, the evolution rate of the released 3DEM confirms a faster exponential growth.
UR - http://www.scopus.com/inward/record.url?scp=85084332962&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85084332962&partnerID=8YFLogxK
U2 - 10.1109/BIBM47256.2019.8983101
DO - 10.1109/BIBM47256.2019.8983101
M3 - Conference contribution
AN - SCOPUS:85084332962
T3 - Proceedings - 2019 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2019
SP - 2661
EP - 2668
BT - Proceedings - 2019 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2019
A2 - Yoo, Illhoi
A2 - Bi, Jinbo
A2 - Hu, Xiaohua Tony
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2019
Y2 - 18 November 2019 through 21 November 2019
ER -
