International Business Machines Corporation

Subhrajit Roy

Stefan Harrer

Benjamin Scott Mashford

Filiz Isabell Kiral-Kornek

Jianbin Tang

Umar Asif

Mahtab Mirmomeni

Susmita Saha

Synaptic reuse allows for a plurality of artificial neurons to be associated with corresponding pluralities of artificial synapses and variable gain amplifiers, to thereby use less space and fewer components to implement and less power to operate than neurons having dedicated paths for each input source. To reduce the likelihood of signal collision, and allow for the independent control and interpretation of input spikes, a router is configured to connect input sources to each of the plurality of artificial neurons in conjunction with a gain configuration controller that is configured to set a gain on each of the plurality of variable gain amplifiers based on a time division schema and an identity of an input source transmitting a spike during a given time.

Artificial intelligence hardware with synaptic reuse

Aspects described herein include a method of generating three-dimensional (3D) spikes. The method comprises receiving a signal comprising time-series data and generating a first two-dimensional (2D) grid. Generating the first 2D grid comprises mapping segments of the time-series data to respective positions of the first 2D grid, and generating, for each position, a spike train corresponding to the respective mapped segment. The method further comprises generating a second 2D grid including performing, for each position, a mathematical operation on the spike train of the corresponding position of the first 2D grid. The method further comprises generating a third 2D grid including performing spatial filtering on the positions of the second 2D grid. The method further comprises generating a 3D grid based on a combination of the first 2D grid, the second 2D grid, and the third 2D grid. The 3D grid comprises one or more 3D spikes.

Generating three-dimensional spikes using low-power computing hardware

One or both of epilepsy seizure detection and prediction at least by performing the following: running multiple input signals from sensors for epilepsy seizure detection through multiple classification models, and applying weights to outputs of each of the classification models to create a final classification output. The weights are adjusted to tune relative output contribution from each classifier model in order that accuracy of the final classification output is improved, while power consumption of all the classification models is reduced. One or both of epilepsy seizure detection and prediction are performed with the adjusted weights. Another method uses streams from sensors for epilepsy seizure detection to train and create the classification models, with fixed weights once trained. Information defining the classification models with fixed weights is communicated to wearable computer platforms for epilepsy seizure detection and prediction. The streams may be from multiple people and applied to an individual person.

Epilepsy seizure detection and prediction using techniques such as deep learning methods

A recommendation method, system, and computer program product include monitoring a patient using a plurality of sensors, receiving patient information including a comfort level corresponding to a sensor of the plurality of sensors, determining a relevance of each sensor of the plurality of sensors to at least one health conditions of the patient, and determining at least one sensor of the plurality of sensors to disconnect based on the comfort level and the relevance of each sensor.

Dynamic recommendations of sensors for patient monitoring

A health-monitoring method, system, and computer program product include operating at least one sensor of a health-monitoring device having a plurality of sensors, detecting a health condition event that requires operation of an additional sensor of the plurality of sensors to monitor the health condition event, activating the additional sensor of the health-monitoring device, and deactivating the additional sensor when the health condition event that requires the second sensor is no longer detected by the detecting.

Energy-efficient multisensory system for mobile health-monitoring

Provided are embodiments for a computer-implemented method. The method includes receiving a sequence of image data, transforming objects in each frame of the sequence of the image data into direction vectors, and clustering the direction vectors based at least in part on features of the objects. The method also includes mapping the direction vectors for the objects in each frame into a position-orientation data structure, and performing tracking using the mapped direction vectors in the position-orientation data structure. Also provided are embodiments of a computer program product and a system for performing object tracking.

Memory-efficient video tracking in real-time using direction vectors

A method for neuromodulation includes monitoring brain activity of a patient using one or more electrodes attached to the patient, and using a first machine learning model to predict whether a patient will have a seizure based on the monitored brain activity of the patient. The method also includes, responsive to the first machine learning model predicting that the patient will have a seizure, using a second machine learning model to determine a neuromodulation signal pattern for preventing the predicted seizure. The method further includes using a neurostimulator to apply the determined neuromodulation signal pattern to the patient. The method also includes, after applying the determined neuromodulation signal pattern to the patient, detecting whether the patient had the predicted seizure based on the monitored brain activity of the patient. The method further includes adjusting at least the second machine learning model based on whether the patient had the predicted seizure.

Seizure detection, prediction and prevention using neurostimulation technology and deep neural network

In an approach, a processor receives classified data, wherein the classified data has been output by a second processor. A processor adjusts a count based on the classified data. A processor determines whether the count is greater than a pre-set threshold, wherein the pre-set threshold is set by a switching module of the processor. Responsive to determining that the count is greater than the pre-set threshold, the processor triggers an alarm of a pre-set alarm length, wherein the pre-set alarm length is set by the switching module of the processor.

Growing community of inventors

Melbourne, Australia

Subhrajit Roy