AI/ML Design & Development
Audio & Image Processing
CROSS PLATFORM Performance Consistency
FDA Regulatory Frameworks
With the rise of technology and computing power in smartphones, personal desktop computers, and cloud-based services, complex Algorithms, Artificial Intelligence (AI) and Machine Learning (ML) have moved rapidly from the realm of theory into actual products with widespread use. Flowbit understands our clients’ interest in the application of these technologies in medical science and engineering. We bring together ideas, techniques, and processes from disparate research papers, journal articles, and books and offer a multi-discipline-based systematic engineering approach.
TensorFlow is an open source software platform consisting of tools, libraries, and community resources for developing and deploying machine learning algorithms. This platform works well with projects requiring access to a significant volume of data. Using this platform, Flowbit can help you build state-of-the-art, scalable and cross-platform machine learning algorithms.
OpenCV — which stands for Open Source Computer Vision Library — provides a large collection of computer vision algorithms, ranging from simple image processing to more complex object detections. These extremely useful algorithms are frequently utilized in combination with our data processing solutions.
Keras is an Application Programming Interface (API), developed in Python for TensorFlow. It allows for ease in developing and training models in Python by offering a consistent and simple API, pre-built functions for common use cases, descriptive logs, and actionable messages.
DIVIDE & CONQUER
DESIGN & DEVELOPMENT
AUDIO & IMAGE PROCESSING
A variety of data formats are collected from all kinds of sensors— for example, camera sensors, oximeter sensors, and temperature sensors. We collect the bytes of information and combine them in the right order, so that the correct picture starts to emerge.
"CL Society 409: Child in a ceremony" by francisco_osorio is licensed with CC BY 2.0.
Depending upon the data processing needs, some data values may need to be normalized, merged, or scaled. This can help with the detection of regions in the data that would otherwise be missed.
A variety of edge detection algorithms can be applied with learned parameters. This helps us get the desired contours for feature detection.
Once all the edges are detected, we can simply filter out the edges we don't want by applying simple threshold or adaptive filters.
Using computer vision, we finally detect the bounding box of the object and therefore, the data in the region of interest that we can further utilize in our analysis.
We have deep expertise in planning and executing verification and validation testing that meets the FDA’s traditional regulatory framework for SaMD. In addition, we work within the new Proposed Regulatory Frameworks by FDA that are under consideration for adoption, such as Algorithm Change Protocol (ACP) in the context of the Total Product Lifecycle (TPL). These and other measures help the FDA to establish reasonable trust of assurance on safety and effectiveness of the medical device, while embracing the iterative improvement power of AI and ML.
AI/ML TPL: KEY NEW ELEMENTS
Thanks to some great AI/ML platforms, such as TensorFlow, CoreML, MATLAB, and PyTorch the actual machine learning code development is only but a small fraction of the effort involved in developing the AI/ML system.
about our approach to algorithms, computer vision, waveform analysis, machine learning engineering best practices as well as anti-patterns. Leave us a question or comment.