Summary
Overview
Work History
Education
Skills
Websites
Publications
Computer Skills And Certificates - Computer Skills
Timeline
Generic

AMIR KONESHLOO

Lubbock

Summary

Experienced data scientist with over seven years in applied mathematics and machine learning, specializing in applying optimization algorithms to solve real world problems.

Overview

8
8
years of professional experience

Work History

Researcher

Self-Learning Skills and Knowledge
06.2023 - Current
  • Conduct cutting-edge research in field advanced Statistics and Probability and Optimization
  • Developed a novel features extraction method using PCA loading plots and Random Forest Regressor to detect the most important features
  • Used Linear regression to present which features impact the target output
  • Used PCA regression to decompose signal visa Hankel matrix and study the features in frequency domain and time domain such as Fast Fourier Transform Entropy, skewness and Kurtosis and observe their impacts on target output
  • Applied and used the Advanced Transformer model 'Autoformer' to successfully predict the next 1 minute of signal
  • This transformer leverages Fast Fourier Transform and the attention mechanism that is the core of the GPT model
  • Actively seek out relevant academic papers, textbooks, and resources to stay abreast of the latest advancements and methodologies in predictive models, probability, and statistics

Postdoctoral Researcher

Texas Tech University
Lubbock
09.2022 - 05.2023
  • Company Overview: Industrial Engineering Dept
  • Analyzed data from burn patients to develop a stronger understanding of over-triage in patient classification
  • Determined criteria for classifying patients as over-triaged or not
  • Identified areas for improvement by considering factors such as burn degree and TBSA
  • Industrial Engineering Dept

Research Data Scientist II

Cleveland Clinic Foundation
Cleveland
08.2021 - 08.2022
  • Found features that indicate the severity of eye disease from eye movement data
  • Used Machine learning and statistics to predict present of eye disease (Amblyopia) from eye movement data
  • Built a data pulling platform using Python to extract data needed for analysis of eye disease
  • Wrote a Query for Teradata in SQL language to locate the distribution blood drawn in the clinic
  • Built a Query for Teradata in SQL language to show the inventory of certain Bio-specimens

Postdoctoral Researcher -Remote

Johns Hopkins Institute for NanoBioTechnology
Baltimore
10.2020 - 07.2021
  • Developed and deployed wearable technologies (Fitbit) for assessing individuals' functional status
  • Analyzed Fitbit usage data for individuals with PAH and correlated it with clinical values using hypothesis testing
  • Utilized Fitbit step count data and K-means clustering to evaluate and group the physical health status of individuals with PAH
  • Assisted in developing patient/clinician apps, APIs (using Flask), and managing databases for remote patient health monitoring
  • The result was published in Nature (npj Digital Medicine)

Research Associate

Stanford University
Palo Alto
01.2020 - 05.2020
  • Company Overview: Pervasive Wellbeing Technology Lab
  • Analyzed and processed time series data to observe changes in physiological measures related to mental health
  • Conducted time series analysis to detect stress from trackpad EDA signals
  • Performed statistical analysis to evaluate the significance of variations in physiological measures, specifically Heart Rate Variability (HRV) features
  • The result of the project was published in the Journal of Medical Internet Research
  • Pervasive Wellbeing Technology Lab

Project Lead and Assistant

Texas Tech University
Lubbock
05.2017 - 12.2019
  • Company Overview: Industrial Engineering Dept
  • Developed a novel cardiac mapping technique using EGM signals for Atrial Fibrillation
  • Identified the atrial fibrillation focal source with 100 % accuracy using statistical techniques and a distributionally robust optimization method applied to EGM
  • Quantified uncertainty associated with measurement noise from input signals and sensor positions for robust source identification
  • Estimated the spatial information of unknown focal sources using regression techniques
  • The result of the project was published in the Bioengineering Journal
  • Developed a heart rate tracking algorithm from wearable devices-NSF Funded Project
  • Pre-processed signals using band-pass filtering, etc., for further analysis
  • Applied joint basis pursuit linear programming for sparse reconstruction of time series
  • Achieved the best output prediction of 2.61 beats per minute (BPM) compared to other existing methods
  • Reduced computational time for efficient online heart rate tracking (only 0.3010 seconds on a regular computer)
  • Developed an automatic predictive model to track heart rate with reduced error
  • The results of the project were published in IEEE Sensors
  • Utilized machine learning techniques for cardiac data analysis
  • Utilized SVM, RF, and NN for accurate region-of-interest identification (above %90 accuracy)
  • Utilized simple linear iterative clustering (SLIC) to reconstruct LGE-MRI images with super-pixels
  • Achieved %96.27 accuracy in heart disease diagnosis using Decision Tree based model - ADABOOST
  • Applied Kernel PCA for improved classification performance
  • Validated classifier performance using K-fold cross-validation
  • The results of the heart disease classification project were published in IISE Annual Conference
  • Industrial Engineering Dept

Graduate Assistant

  • Mentored a group of Industrial Engineers in Reliability Theory, Risk Modeling Assessment, and Deterministic Optimization
  • Guided Industrial Engineers on mathematical problem-solving projects
  • Supervised a group of Industrial Engineers in Deterministic Operations Research

Education

Ph.D. - Industrial Engineering

Texas Tech University
Lubbock, TX
08.2020

M.S. - Industrial Engineering

Wichita State University
Wichita, KS
05.2015

Skills

  • Optimization
  • Predictive Modeling
  • Generative Models
  • Statistical Inference
  • Statistical Analysis
  • Deep Learning
  • Deep Reinforcement Learning
  • ML Deployment

Publications

  • Https://scholar.google.com/citations?user=wCJ7gLwAAAAJ
  • P. Searson, Z. Xu, N. Zahradka, S. Ip, A. Koneshloo, R. Roemmich, S. Sehgal, and K. Highland, "Evaluation of Physical Health Status Beyond Daily Step Count Using a Wearable Activity Sensor", npj Digital Medicine, 2022.
  • R. Goel, M. An, H. Alayrangues, A. Koneshloo, E. Lincoln, P. Paredes, "Stress Tracker-Detecting Acute Stress From a Trackpad: Controlled Study", Journal of Medical Internet Research, 2020.
  • A. Koneshloo, D. Du, Y. Du, "An Uncertainty Modeling Framework for Intracardiac Electrogram Analysis", Bioengineering Journal, 2020.
  • A. Koneshloo, D. Du, "A Novel Motion Artifact Removal Method Via Joint Basis Pursuit Linear Program to Accurately Monitor Heart Rate", IEEE Sensors Journal, 2019.
  • A. Koneshloo, D. Du, "Coronary Heart Disease Diagnosis Using Kernel PCA and Adaptive Boosting", IISE Proceedings, 2018.

Computer Skills And Certificates - Computer Skills

  • Python
  • SQL
  • MATLAB
  • GraphPad
  • ARENA

Timeline

Researcher

Self-Learning Skills and Knowledge
06.2023 - Current

Postdoctoral Researcher

Texas Tech University
09.2022 - 05.2023

Research Data Scientist II

Cleveland Clinic Foundation
08.2021 - 08.2022

Postdoctoral Researcher -Remote

Johns Hopkins Institute for NanoBioTechnology
10.2020 - 07.2021

Research Associate

Stanford University
01.2020 - 05.2020

Project Lead and Assistant

Texas Tech University
05.2017 - 12.2019

Graduate Assistant

Ph.D. - Industrial Engineering

Texas Tech University

M.S. - Industrial Engineering

Wichita State University
AMIR KONESHLOO