Scientist to Use AI For Newborns Diagnostics

The main goal of the study is to create an algorithm that detects deviations in the development of limb movements of newborns in the first few months
12 July 2018   768

A team of scientists from the University of Southern California and the University of Madrid used AI to detect abnormalities in the development of newborns. The algorithm classifies the movements of the limbs and according to these data creates a forecast is for 1-12 months. This is reported by Venture Beat.

Scientists used the data of the laboratory for monitoring neuromotorics of newborns, located at the University of Southern California. Accelerometers, gyroscopes and magnetometers were attached to the feet of children. The algorithm collected data from the sensors for the left and right legs, then calculated the duration of the movements, the average and maximum acceleration, and other indicators.

Then the developers manually entered the age of the child, a scaled development score and information about it (typical or atypical), collected the predictive model. After using binary classification algorithms, taking into account the 3 best results for minimizing errors.

Based on the obtained data, artificial intelligence predicted delays in development for the first six months with an accuracy of 83.9%. For a period of 6-12 months, the accuracy was slightly lower - 77%. Detailed text and results are published in the article.

[S]tudies have demonstrated that kinematic variables, such as kicking frequency, spatiotemporal organization, and interjoint and interlimb coordination, are different between infants with typical development … and infants at risk … including infants with intellectual disability, myelomeningocele, Down syndrome, as well as infants born preterm.


The main goal of the study is to create an algorithm that detects deviations in the development of limb movements of newborns in the first few months. This will allow to take purposeful actions. Studies have shown that between children with normal development and children in the risk group, there are kinematic differences. The latter include the frequency of movement of the legs, spatial orientation and coordination of the limbs.

Nvidia to Open SPADE Source Code

SPADE machine learning system creates realistic landscapes based on rough human sketches
15 April 2019   673

NVIDIA has released the source code for the SPADE machine learning system (GauGAN), which allows for the synthesis of realistic landscapes based on rough sketches, as well as training models associated with the project. The system was demonstrated in March at the GTC 2019 conference, but the code was published only yesterday. The developments are open under the non-free license CC BY-NC-SA 4.0 (Creative Commons Attribution-NonCommercial-ShareAlike 4.0), allowing use only for non-commercial purposes. The code is written in Python using the PyTorch framework.

Sketches are drawn up in the form of a segmented map that determines the placement of exemplary objects on the scene. The nature of the generated objects is set using color labels. For example, a blue fill turns into sky, blue into water, dark green into trees, light green into grass, light brown into stones, dark brown into mountains, gray into snow, a brown line into a road, and a blue line into the river. Additionally, based on the choice of reference images, the overall style of the composition and the time of day are determined. The proposed tool for creating virtual worlds can be useful to a wide range of specialists, from architects and urban planners to game developers and landscape designers.

Objects are synthesized by a generative-adversarial neural network (GAN), which, based on a schematic segmented map, creates realistic images by borrowing parts from a model previously trained on several million photographs. In contrast to the previously developed systems of image synthesis, the proposed method is based on the use of adaptive spatial transformation followed by transformation based on machine learning. Processing a segmented map instead of semantic markup allows you to achieve an exact match of the result and control the style.

To achieve realism, two competing neural networks are used: the generator and the discriminator (Discriminator). The generator generates images based on mixing elements of real photos, and the discriminator identifies possible deviations from real images. As a result, a feedback is formed, on the basis of which the generator begins to assemble more and more qualitative samples, until the discriminator ceases to distinguish them from the real ones.