Machine Learning Postdoc Fellow

Our laboratory applies computational and machine learning methods to investigate the impact of seizures and abnormal brain activity on outcomes in pigs with cortical impact. Our goal is to understand pathological correlates of epilepsy and traumatic brain injury. Analysis of datasets (including video-EEG telemetry, intracellular Chloride, among others) is central to these efforts.

Specific efforts focus on developing methods for automatically classifying the semiology of pigs in video monitoring as they undergo the development of epilepsy and understanding the relationships between any abnormal behaviors and time after injury or the change in seizure frequency. Efforts will particularly focus on using supervised machine learning approaches including training artificial neural networks via open source software such as Keras, Tensorflow, DeepLabCut, SimBA, TREBA etc. or unsupervised learning methods, heuristics, and other algorithms to learn patterns, fit and extrapolate from models, and process large datasets of video frames.

The person will interact with staff in other lab's such as Sydney Cash's lab, Kevin Staley's lab, and Kyle Lillis' lab.

PRINCIPAL DUTIES AND RESPONSIBILITIES:

The machine learning engineer will work and mentor a team of researchers in searching for patterns hidden in large data sets for research in neurology. The machine learning engineer will be responsible for data from the electronic data repository, including EEG, video, and peripheral blood biomarkers. The machine learning engineer will develop unique algorithmic approaches for analysis of data and supervise and mentor a team of research staff. Responsibilities will include:

- Creating or applying methods for automatic classification or regression on large data

- Software development and code management

- Data wrangling of biological, instrumental, or technical data

- Guiding a team on computational tasks and helping oversee research staff

- Problem solving and troubleshooting of technical problems for research staff

- Management of a large physiological database, warehouse, and/or repository

- Development of algorithms and maintaining a software pipeline

- Collaborate and interface with personnel from other research laboratories

- Documenting steps for reproducing results

- Outlining desired milestones for research staff so that objectives can be met

- Generate reports of statistical analysis

- Prepare and submit research manuscripts and abstracts

- Provide weekly updates on data processing, analysis or other research progress

- Present at lab meetings, and at local and national meetings

- Data annotation, storage, and management

- Communicating concepts in a helpful way to those that are not computer scientists

* Excellent analytical and troubleshooting skills,

* Demonstrated knowledge of software development methodologies and software pipeline design.

* Ability to solve complex and large-scale problems to make important contributions to medicine and science.

* Strong software engineering and quantitative background including knowledge in Python, Unix Shell (e.g. Bash, Zsh, etc), deep neural networks of different architectures (convolutional, recurrent, etc), algorithms (sorting, binary search, etc), C++ or any other compiler based language(s), calculus, basic statistics (hypothesis testing, distributions, regression, etc), data visualization, etc.

* Experience in scientific method and critical thinking

* Detail-oriented and pro-active workstyle

* Strong ethical principles

* Ability to work independently and as part of a team

* Excellent verbal and written communication skills

Any additional skills are a plus including:

Parallel computing, command prompt, working with GPUs, supercomputing, video software (e.g. ffmpeg), SQL, large language models, MATLAB, PHP, or other additional languages, hardware knowledge, advanced understanding of kernel, neurobiology knowledge, et

EDUCATION:

PhD in a relevant discipline such as: computer science, math, computer engineering, statistics, cognitive science, electrical engineering, bioengineering, data science, etc.

EXPERIENCE: Specify minimum creditable years of experience and clearly indicate if preferred or required

Minimum of 2 years of relevant experience required.; Knowledge of some Computer Science/Engineering concepts required.

FISCAL RESPONSIBILITY:

No budgetary responsibility but will need to design projects and identify equipment for projects within a budgetary scope and in liaison with a staff assistant performing purchase orders.

WORKING CONDITIONS:

Work will be performed in CNY Building 149 and some work on and in laboratory space on the MGH main campus.