Abstract: In a method of forming a pattern, a first pattern is formed over an underlying layer, the first pattern including main patterns and a lateral protrusion having a thickness of less than 25% of a thickness of the main patterns, a hard mask layer is formed over the first pattern, a planarization operation is performed to expose the first pattern without exposing the lateral protrusion, a hard mask pattern is formed by removing the first pattern while the lateral protrusion being covered by the hard mask layer, and the underlying layer is patterned using the hard mask pattern as an etching mask.

Abstract: An eye tracking device and an eye tracking method are provided. The eye tracking device includes a sensor and multiple signal transmitting units. The sensor is disposed on an eyeball of a user. The sensor has multiple signal receiving units. The signal transmitting units are disposed around the sensor and surround the sensor. The signal transmitting units respectively transmit multiple transmitting signals. The signal receiving units receive the transmitting signals to respectively generate multiple sensing signals. The eye tracking device calculates a position of the eyeball according to the sensing signals.

Abstract: A first group of semiconductor fins are over a first region of a substrate, the substrate includes a first stepped profile between two of the first group of semiconductor fins, and the first stepped profile comprises a first lower step, two first upper steps, and two first step rises extending from opposite sides of the first lower step to the first upper steps. A second group of semiconductor fins are over a second region of the substrate, the substrate includes a second stepped profile between two of the second group of semiconductor fins, and the second stepped profile comprises a second lower step, two second upper steps, and two second step rises extending from opposite sides of the second lower step to the second upper steps, in which the second upper steps are wider than the first upper steps in the cross-sectional view.

Abstract: A control device and a correcting method of a control device with a strength feedback function are disclosed. The control device includes a vibrator, a driver, a strength sensor, and a processor. In correction procedures, the driver provides a driving signal with a plurality of different setting voltage values respectively to the vibrator. In each correction procedure, the driver adjusts a frequency of the driving signal to setting frequency values by time division. In the correction procedures, the strength sensor senses a strength generated by a vibration of the vibrator to obtain strength sensing information respectively corresponding to the setting voltage values. The processor analyzes the strength sensing information, the setting frequency values, and the setting voltage values to obtain relationship information between the setting frequency values and the setting voltage values corresponding to each strength sensing information.

Abstract: An eye tracking device and an eye tracking method are provided. The eye tracking device includes a sensor and multiple signal transmitting units. The sensor is disposed on an eyeball of a user. The sensor has multiple signal receiving units. The signal transmitting units are disposed around the sensor and surround the sensor. The signal transmitting units respectively transmit multiple transmitting signals. The signal receiving units receive the transmitting signals to respectively generate multiple sensing signals. The eye tracking device calculates a position of the eyeball according to the sensing signals.

Abstract: A stress feedback device and a virtual image display system are provided. The stress feedback device includes a position sensor and a pressure generator. The position sensor is configured to generate a position information based on a trigger signal and send the position information to a host. The pressure generator is configured to apply a stress to a target area based on the trigger signal and adjust an intensity of the stress.

Abstract: A method for predicting air quality with the aid of machine learning models includes: (A) providing air pollution data to perform an eXtreme Gradient Boosting (XGBoost) regression algorithm for obtaining a XGBoost prediction value; (B) providing the air pollution data to perform a Long Short-Term Memory (LSTM) algorithm for obtaining an LSTM prediction value; (C) combining the air pollution data, the XGBoost prediction value and the LSTM prediction value to generate air pollution combination data; (D) performing an XGBoost classification algorithm to obtain a suggestion for whether to issue an air pollution alert; and (E) performing the XGBoost regression algorithm on the air pollution combination data to obtain an air pollution prediction value. Two layers of machine learning models are built, and a situation where prediction results are too conservative when a single model does not have enough data can be improved.

Abstract: A logarithmic power detector includes a power distributor, a first detection circuit, a second detection circuit and an output circuit. The power distributor is used to generate a first power signal and a second power signal according to an input signal. The first detection circuit is used to attenuate the first power signal to generate a first rectified signal, filter the first rectified signal to generate a first low-pass signal, and amplify the first low-pass signal to generate a first amplification current. The second detection circuit is used to attenuate the second power signal to generate a second rectified signal, filter the second rectified signal to generate a second low-pass signal, and amplify the second low-pass signal to generate a second amplification current. The output circuit is used to receive the first amplification current and the second amplification current to generate a converted voltage related to the input signal.

Abstract: A spliced unit including a substrate, a circuit unit, and light-emitting units is provided. The substrate includes a first part having a first bottom surface and a first top surface opposite to the first bottom surface, and a second part having a second bottom surface and a second top surface opposite to the second bottom surface. There is a height difference between the first bottom surface of the first part and the second bottom surface of the second part. The circuit unit is disposed on the first top surface. The light-emitting units are disposed in the second part of the substrate. In a direction of a normal line perpendicular to the substrate, the first part of the substrate and the second part of the substrate are not overlapped, and the circuit unit and the light-emitting units are not overlapped. A spliced panel including the spliced units is also provided.

Abstract: A logarithmic power detector includes a power distributor, a first detection circuit, a second detection circuit and an output circuit. The power distributor is used to generate a first power signal and a second power signal according to an input signal. The first detection circuit is used to attenuate the first power signal to generate a first rectified signal, filter the first rectified signal to generate a first low-pass signal, and amplify the first low-pass signal to generate a first amplification current. The second detection circuit is used to attenuate the second power signal to generate a second rectified signal, filter the second rectified signal to generate a second low-pass signal, and amplify the second low-pass signal to generate a second amplification current. The output circuit is used to receive the first amplification current and the second amplification current to generate a converted voltage related to the input signal.

Abstract: An anti-human T-cell immunoglobulin domain and mucin domain 3 (TIM-3) antibody, can bind the peptides, comprising the amino-acid sequence RKGDVSL (SEQ ID NO: 9) and/or EKFNLKL (SEQ ID NO: 10) of human TIM-3 protein. The antibody can regulate immune cell activity. The antibody or binding fragment thereof is useful in diagnosis, prognosis, and treatment of cancers that have been reported to express cell-surface TIM-3 such as lung, liver, esophageal cancer and solid tumors.

Abstract: A spliced unit including a substrate, a circuit unit, and light-emitting units is provided. The substrate includes a first part having a first bottom surface and a first top surface opposite to the first bottom surface, and a second part having a second bottom surface and a second top surface opposite to the second bottom surface. There is a height difference between the first bottom surface of the first part and the second bottom surface of the second part. The circuit unit is disposed on the first top surface. The light-emitting units are disposed in the second part of the substrate. In a direction of a normal line perpendicular to the substrate, the first part of the substrate and the second part of the substrate are not overlapped, and the circuit unit and the light-emitting units are not overlapped. A spliced panel including the spliced units is also provided.

Abstract: A method for predicting air quality with the aid of machine learning models includes: (A) providing air pollution data to perform an eXtreme Gradient Boosting (XGBoost) regression algorithm for obtaining a XGBoost prediction value; (B) providing the air pollution data to perform a Long Short-Term Memory (LSTM) algorithm for obtaining an LSTM prediction value; (C) combining the air pollution data, the XGBoost prediction value and the LSTM prediction value to generate air pollution combination data; (D) performing an XGBoost classification algorithm to obtain a suggestion for whether to issue an air pollution alert; and (E) performing the XGBoost regression algorithm on the air pollution combination data to obtain an air pollution prediction value. Two layers of machine learning models are built, and a situation where prediction results are too conservative when a single model does not have enough data can be improved.

Abstract: The present invention relates to hydrogenated block copolymers of vinyl aromatic polymer blocks and conjugated diene polymer blocks having specific molecular weights, compositions, molecular structures and architectures, such that improved processability, mechanical and optical properties are attained. The hydrogenated block copolymer can be further added with different hydrogenated block copolymers to enhance the desired properties.

Abstract: A display panel includes waveguides, wires and a pixel array. The pixel array includes a plurality of pixel units. The pixel units are arranged in a plurality of columns and a plurality of rows. Each pixel unit includes a pixel electrode, a light filtering unit, and a photo transistor. The light filtering unit is coupled to one of the waveguides. The photo transistor is electrically connected to the pixel electrode and one of the wires, and is coupled to the light filtering unit. The waveguide transmits a light control signal. Each wire transmits an electric control signal. The light filtering unit is configured to receive a sub control signal from the waveguides to which the light filtering unit is coupled and filter out a specific optical signal according to the received sub control signal as an input signal of the photo transistor.

Abstract: The present invention relates to hydrogenated block copolymers of vinyl aromatic polymer blocks and conjugated diene polymer blocks having specific molecular weights, compositions, molecular structures and architectures, such that improved processability, mechanical and optical properties are attained. The hydrogenated block copolymer can be further added with different hydrogenated block copolymers to enhance the desired properties.

Abstract: An anti-human T-cell immunoglobulin domain and mucin domain 3 (TIM-3) antibody, can bind the peptides, comprising the amino-acid sequence RKGDVSL (SEQ ID NO: 9) and/or EKFNLKL (SEQ ID NO: 10) of human TIM-3 protein. The antibody can regulate immune cell activity. The antibody or binding fragment thereof is useful in diagnosis, prognosis, and treatment of cancers that have been reported to express cell-surface TIM-3 such as lung, liver, esophageal cancer and solid tumors.

Abstract: A display panel includes waveguides, wires and a pixel array. The pixel array includes a plurality of pixel units. The pixel units are arranged in a plurality of columns and a plurality of rows. Each pixel unit includes a pixel electrode, a light filtering unit, and a photo transistor. The light filtering unit is coupled to one of the waveguides. The photo transistor is electrically connected to the pixel electrode and one of the wires, and is coupled to the light filtering unit. The waveguide transmits a light control signal. Each wire transmits an electric control signal. The light filtering unit is configured to receive a sub control signal from the waveguides to which the light filtering unit is coupled and filter out a specific optical signal according to the received sub control signal as an input signal of the photo transistor.

Abstract: The invention relates to a method for enhancing wound healing in a subject in need thereof, comprising administering to the subject a composition comprising fibronectin type III domain-containing protein 5 (FNDC5) or its cleaved fragment irisin in an amount effective to enhance wound healing

Abstract: A pace speed advising method, a pace speed advising apparatus, and a recording medium using the method are provided. In the method, a total target distance and a total target time of a user's path are obtained, and separated into multiple time slots. Then, a route of the user's path is obtained and a route information of the route is retrieved. Finally, a pace speed of each time slot is advised according to the route information.