Abstract: Systems and methods are provided for enhanced heart medical imaging operations, particularly as by incorporating use of artificial intelligence (AI) based fetal heart functional analysis and/or real-time and automatic ejection fraction (EF) measurement and analysis.

Abstract: Methods and systems are provided for augmenting ultrasound image training data, which may be used to train one or more machine learning models. One example method for augmenting ultrasound training data comprises, selecting an ultrasound image and a ground truth output corresponding to the ultrasound image, determining a first modification to apply to the ultrasound image, applying the first modification to the ultrasound image to produce an augmented ultrasound image, modifying the ground truth output based on the first modification to produce an augmented ground truth output corresponding to the augmented ultrasound image, and training a machine learning model using the augmented ultrasound image and the augmented ground truth output. In this way, a machine learning model may learn a more robust mapping from ultrasound image features to expected output, with less probability of overfitting, and with increased generalizability to noisy ultrasound images, or ultrasound images containing artifacts.

Abstract: Systems and methods are provided for enhanced heart medical imaging operations, particularly as by incorporating use of artificial intelligence (AI) based fetal heart functional analysis and/or real-time and automatic ejection fraction (EF) measurement and analysis.

Abstract: Methods and systems are provided for augmenting ultrasound image training data, which may be used to train one or more machine learning models. One example method for augmenting ultrasound training data comprises, selecting an ultrasound image and a ground truth output corresponding to the ultrasound image, determining a first modification to apply to the ultrasound image, applying the first modification to the ultrasound image to produce an augmented ultrasound image, modifying the ground truth output based on the first modification to produce an augmented ground truth output corresponding to the augmented ultrasound image, and training a machine learning model using the augmented ultrasound image and the augmented ground truth output. In this way, a machine learning model may learn a more robust mapping from ultrasound image features to expected output, with less probability of overfitting, and with increased generalizability to noisy ultrasound images, or ultrasound images containing artifacts.

Abstract: An apparatus may include an image sensor that contains a multiplicity of pixel elements to detect one or more images and a processor circuit coupled to the image sensor. The apparatus may include a white balance module for execution on the processor circuit to receive, based upon a detected image of the one or more images, for a plurality of pixel elements of the multiplicity of pixel elements, three of more gray level values for a respective three or more color channels, to determine grayness likelihood functions for the respective three or more color channels, the three or more grayness likelihood functions comprising a proportional contribution to grey pixels of the detected image from one or more gray levels for each respective color channel, and to determine a white balance gain for two or more color channels based upon the determined grayness likelihood functions. Other embodiments are described and claimed.

Abstract: An apparatus may include an image sensor that contains a multiplicity of pixel elements to detect one or more images and a processor circuit coupled to the image sensor. The apparatus may include a white balance module for execution on the processor circuit to receive, based upon a detected image of the one or more images, for a plurality of pixel elements of the multiplicity of pixel elements, three of more gray level values for a respective three or more color channels, to determine grayness likelihood functions for the respective three or more color channels, the three or more grayness likelihood functions comprising a proportional contribution to grey pixels of the detected image from one or more gray levels for each respective color channel, and to determine a white balance gain for two or more color channels based upon the determined grayness likelihood functions. Other embodiments are described and claimed.

Abstract: Outdoor imaging is plagued by poor visibility conditions due to atmospheric scattering, particularly in haze. A major problem is spatially-varying reduction of contrast by stray radiance (airlight), which is scattered by the haze particles towards the camera. The images can be compensated for haze by subtraction of the airlight and correcting for atmospheric attenuation. Airlight and attenuation parameters are computed by analyzing polarization-filtered images. These parameters were estimated in past studies by measuring pixels in sky areas. However, the sky is often unseen in the field of view. The invention provides methods for automatically estimating these parameters, when the sky is not in view.