Abstract: Provided is a vehicle thermal management system, including: an HVAC subsystem thermally connected to a passenger compartment; and a powertrain cooling subsystem thermally connected to a powertrain component.

Abstract: An active noise control system selects one reference sensor providing a reference signal with the largest coherence to a noise signal, among a plurality of available reference sensors, as a first entry of a reference sensor set. After selecting the first entry, the active noise control system repeats a process in which a sensor capable of providing the largest information quantity to a current reference sensor set among remaining sensors is selected as a new entry of a reference sensor set, until a desired number of sensors is reached or a desired control level is reached. When the reference sensor set is determined, the active noise control system utilizes the entries of the reference sensor set to generates a noise control signal suitable for canceling the noise signal.

Abstract: Supervised and unsupervised learning schemes may be used to automatically label medical images for use in deep learning applications. Large labeled datasets may be generated from a small initial training set using an iterative snowball sampling scheme. A machine learning powered automatic organ classifier for imaging datasets, such as CT datasets, with a deep convolutional neural network (CNN) followed by an organ dose calculation is also provided. This technique can be used for patient-specific organ dose estimation since the locations and sizes of organs for each patient can be calculated independently.

Abstract: A method for generating a synthesized medical image by receiving a normal image includes generating first data based on a random selection, generating second data, and, based at least in part on the first and second data, modifying the normal image to form the synthesized medical image. Modifying the normal image comprises combining the first data and the second data. The first data characterizes an image that represents a lesion and the second data characterizes a transformation of that image as well as a location of the lesion.

Abstract: A method includes receiving, from a patient, an image having a visible lesion, modifying the image to appear as if the lesion were not present, thereby forming a second image, generating a delineation of the abnormality using a difference between the first and second images, and tagging the segmented lesions.

Abstract: This invention relates to estimating the window width and window level (center) which are typically used to view and then transform diagnostic imaging data to grayscale images. These grayscale images are then used to check the presence of diseases or abnormalities. For each individual diagnostic image, this invention automatically estimates the most appropriate values. This automatic estimation is done by a specialized module added on to a convolutional neural network-based disease detection system.

Abstract: A display device includes a timing controller which supplies a clock training signal through a data clock signal line and a first control signal through a shared signal line in a first period of one frame, and supplies image data through the data clock signal line in a second period of the one frame, a data driver provided with data driving circuits which generate a clock signal based on the clock training signal and the first control signal in the first period, and generate data voltages based on the clock signal and the image data in the second period, and a pixel part which receives the data voltages from the data driver. The data driver may supply a second control signal indicating a reception state of the data driver to the timing controller through the shared signal line in the second period.

Abstract: A display device includes a timing controller which supplies a clock training signal through a data clock signal line and a first control signal through a shared signal line in a first period of one frame, and supplies image data through the data clock signal line in a second period of the one frame, a data driver provided with data driving circuits which generate a clock signal based on the clock training signal and the first control signal in the first period, and generate data voltages based on the clock signal and the image data in the second period, and a pixel part which receives the data voltages from the data driver. The data driver may supply a second control signal indicating a reception state of the data driver to the timing controller through the shared signal line in the second period.

Abstract: A display panel driving apparatus includes a data driving part, a data driving part and an off voltage controlling part. The data driving part is configured to output a data signal to a data line of a display panel. The gate driving part is configured to output a gate signal to a gate line of the display panel. The off voltage controlling part is configured to receive a first off voltage and a second off voltage applied to the gate driving part to generate the gate signal, measure a leakage current of the gate driving part, and control the first off voltage based on the leakage current. Thus, display quality of a display apparatus including the gate driving part may be improved.

Abstract: A gate driving circuit includes a shift register configured to generate a plurality of output signals based on at least one clock signal, a plurality of output buffers configured to generate a plurality of gate signals by amplifying the output signals and to sequentially output the gate signals to a plurality of gate lines in a display panel, a detector configured to sequentially sense the gate signals and to compare each of the gate signals to a reference voltage, and a dummy output buffer configured to be coupled between the shift register and a gate line of the gate lines instead of an output buffer of the output buffers when a voltage level of a corresponding gate signal from the output buffers is less than a voltage level of the reference voltage.

Abstract: A display device includes a timing controller configured to supply a first set signal to a data control signal line in a first frequency mode, and to supply a second set signal and a data signal to the data control signal line in a second frequency mode that is different from the first frequency mode, a data driver configured to recover the data signal supplied to the data control signal line according to a signal recovery characteristic value, to generate a plurality of data voltages based on the recovered data signal, and to adjust the signal recover characteristic value based on the first set signal and the second set signal, and a display unit including a plurality of pixels that emit lights with gray scales corresponding to the plurality of data voltages.

Abstract: A display device includes a display panel including data lines, gate lines, and pixels connected to the data lines and the gate lines, a data driver configured to drive the data lines, a plurality of gate drivers, each configured to drive a corresponding portion of the gate lines, and a plurality of feedback lines connected to one of the data lines at a plurality of measurement positions corresponding to the plurality of gate drivers. The data driver applies a test voltage to the data line, receives the test voltage as a plurality of feedback voltages through the plurality of feedback lines, and determines gate shift amounts at the plurality of measurement positions corresponding to the plurality of gate drivers based on the plurality of feedback voltages. The gate drivers apply gate signals to the gate lines that are shifted by the determined gate shift amounts.

Abstract: Supervised and unsupervised learning schemes may be used to automatically label medical images for use in deep learning applications. Large labeled datasets may be generated from a small initial training set using an iterative snowball sampling scheme. A machine learning powered automatic organ classifier for imaging datasets, such as CT datasets, with a deep convolutional neural network (CNN) followed by an organ dose calculation is also provided. This technique can be used for patient-specific organ dose estimation since the locations and sizes of organs for each patient can be calculated independently.

Abstract: A display device may include a timing controller, a data driver and a plurality of pixels. The timing controller supplies a clock training pattern over a data/clock signal line in a first time period, and supplies pixel/control data over the data/clock signal line in a second time period. The data driver generates a clock signal, using the clock training pattern, in the first period, and generate a plurality of data voltages based on the plurality of pixel data, using the clock signal, in the second period. The plurality of pixels receive the plurality of data voltages and emit corresponding light. During the second period, the data driver outputs a feedback signal to the timing controller indicating that the locking of the clock signal has failed. The timing controller re-supplies the clock training pattern in response to the feedback signal.

Abstract: A display device includes a display panel, a source driver, and an alignment detection circuit. The display panel includes data lines and pads connected to the data lines. The source driver includes output lines connected to the pads to supply a data signal, and a detection circuit to selectively connect a first detection line supplied with a first detection voltage and a second detection line supplied with a second detection voltage to the output lines. The alignment detection circuit includes a detection capacitor connected between the first detection line and the second detection line, and a voltage detection circuit connected to one end of the detection capacitor to detect a voltage of the detection capacitor. The detection circuit connects a (2n?1)-th output line of the output lines to the first detection line, and connects a (2n)-th output line of the output lines to the second detection line.

Abstract: This invention relates to estimating the window width and window level (center) which are typically used to view and then transform diagnostic imaging data to grayscale images. These grayscale images are then used to check the presence of diseases or abnormalities. For each individual diagnostic image, this invention automatically estimates the most appropriate values. This automatic estimation is done by a specialized module added on to a convolutional neural network-based disease detection system.

Abstract: A gate driving circuit includes a shift register configured to generate a plurality of output signals based on at least one clock signal, a plurality of output buffers configured to generate a plurality of gate signals by amplifying the output signals and to sequentially output the gate signals to a plurality of gate lines in a display panel, a detector configured to sequentially sense the gate signals and to compare each of the gate signals to a reference voltage, and a dummy output buffer configured to be coupled between the shift register and a gate line of the gate lines instead of an output buffer of the output buffers when a voltage level of a corresponding gate signal from the output buffers is less than a voltage level of the reference voltage.

Abstract: A method for generating a synthesized medical image by receiving a normal image includes generating first data based on a random selection, generating second data, and, based at least in part on the first and second data, modifying the normal image to form the synthesized medical image. Modifying the normal image comprises combining the first data and the second data. The first data characterizes an image that represents a lesion and the second data characterizes a transformation of that image as well as a location of the lesion.

Abstract: A display device includes a display panel including data lines, gate lines, and pixels connected to the data lines and the gate lines, a data driver configured to drive the data lines, a plurality of gate drivers, each configured to drive a corresponding portion of the gate lines, and a plurality of feedback lines connected to one of the data lines at a plurality of measurement positions corresponding to the plurality of gate drivers. The data driver applies a test voltage to the data line, receives the test voltage as a plurality of feedback voltages through the plurality of feedback lines, and determines gate shift amounts at the plurality of measurement positions corresponding to the plurality of gate drivers based on the plurality of feedback voltages. The gate drivers apply gate signals to the gate lines that are shifted by the determined gate shift amounts.

Abstract: A display device includes a timing controller configured to supply a first set signal to a data control signal line in a first frequency mode, and to supply a second set signal and a data signal to the data control signal line in a second frequency mode that is different from the first frequency mode, a data driver configured to recover the data signal supplied to the data control signal line according to a signal recovery characteristic value, to generate a plurality of data voltages based on the recovered data signal, and to adjust the signal recover characteristic value based on the first set signal and the second set signal, and a display unit including a plurality of pixels that emit lights with gray scales corresponding to the plurality of data voltages