Open Brain AI Timeline
About Open Brain AI
Open Brain AI
Open Brain AI is a research platform. We offer computational solutions for the automatic analysis of spoken and written discourse and provide informed computational measures for diagnosis, prognosis, treatment efficacy assessment, and motivated awareness of speech, language, and communication impairments and related conditions. We mainly focus on aphasia, dementia, and developmental language conditions.
Aim
Our goal is to prioritize user-friendliness and accessibility in the most seamless manner possible. We aim to eliminate the need for users to download any software, write code, or grapple with computational complexities. We also recognize the importance of offering a robust, precise, and adaptable environment that generates versatile outputs. This allows clinicians and researchers to seamlessly integrate the output into their existing toolset, including familiar software like Excel or other statistical analysis programs.
AI Analysis of Discourse
We provide discourse analysis using AI. We offer recommendations on the Cohesion and Coherence, checks the errors, and suggests whether there are pathological impairments.
Clinical Language Assessment:
Written Language
We analyze written language and provide measures and scores about all linguistic domains, including phonology, morphology, syntax, semantics, lexicon, and readability measures.
Clinical Language Assessment:
Spoken Language
Automatic transcription of speech and scoring of linguistic domains, including phonology, morphology, syntax, semantics, lexicon, and readability measures.
Clinical Toolkit
Automatic sound recording analysis, semantics scoring, spelling performance, and phonological performace.
Open Brain AI - 15 years of research.
A birdseye view of Open Brain AI main developments.
Open Brain AI is an advanced computational platform and a sophisticated ensemble of tools that have been meticulously developed and refined as early as 2007.
-
Early Years.
The first code was developed for Charalambos Themistocleous PhD research in 2007. He developed text to IPA converter and early text-to-sound alignment facilitate his research.
-
2015-2018: Themis
Speech-to-Text transcription interfaces for Greek, English, and Swedish. The release of Tensorflow by Google allowed us to quicly develop and test various types of Neural Nets, including Recurrent (LSTM), Convolutional, and Feedforward Deep Neural Networks. We build the a machine learning application for the identification of patients with Mild Cognitive Impairment from Healthy Controls.
-
Putting Everything Together - 2018-2020
Developing a pipeline for analyzing audio recordings of patients, transcribing them automatically, and conducting measures of grammar and speech. Subsequently the measures are feeded to Deep Neural Networks to provide diagnosis. We applied the system to subtype patients with Primary Progressive Aphasia.
-
2020 - Website Development
In the shadow of the global pandemic in 2020, Open Brain AI evolved into a unified interface for these diverse tools, breathing new life into them.
The Django site was developed and the Domain Name Open Brain AI was acquired in a cafe under the Acropolis in Athens, Greece. Open Brain AI was underwent further refinements in the summer that followed. A spelling algorithm, the subject of a previously published paper co-authored with Brenda Rapp, constitutes another vital component. The source code of the spelling up is available on Github.
