Abstract: The present application discloses a backlight module and a display apparatus with the backlight module. The backlight module includes a substrate, a plurality of light sources, and a plurality of reflectors. The reflectors are arranged on the peripheries of the light sources. The surfaces, facing the light sources, of the reflectors are multi-stage curved surfaces. Light emitted from a side surface of the light source basically emits in the vertical direction after being adjusted by the multi-stage curved surface of the reflector, so as to collimate the light emitted from the light source, prevent the light emitted from the lighted light source on the periphery from irradiating an unlighted area nearby, and prevent light leakage. By using the reflector with this structure, in internal space formed in the reflector, the light in the internal space formed by each reflector is uniform, improving the optical grade of the display apparatus.

Abstract: A surface modifying apparatus is configured to modify a bonding surface of a substrate to be bonded to another substrate by plasma of a processing gas. The surface modifying apparatus includes a processing vessel; a measuring unit; and a controller. The processing vessel is configured to accommodate the substrate therein. The measuring unit is configured to measure a value indicating an amount of moisture in the processing vessel. The controller is configured to determine whether or not bonding strength between the substrate and the another substrate, when it is assumed that the substrate modified in the processing vessel is bonded to the another substrate, is good based on the value indicating the amount of moisture in the processing vessel measured by the measuring unit.

Abstract: A surface modifying method of modifying a bonding surface of a substrate to be bonded to another substrate by plasma of a processing gas includes an adjusting process and a modifying process. In the adjusting process, an amount of moisture in a processing vessel is adjusted by supplying a humidified gas into the processing vessel allowed to accommodate the substrate therein. In the modifying process, the bonding surface of the substrate is modified by forming the plasma of the processing gas in the processing vessel in a state that the amount of moisture in the processing vessel is adjusted.

Abstract: Provided is a device performing machine learning by extracting an acoustic feature value from conversational voice data and predicting a disease level of a subject on the basis of a disease prediction model to be generated by the machine learning, the device including: a matrix calculation unit 23 calculating a spatial delay matrix using a relation value of a plurality of types of acoustic feature values; and a matrix decomposition unit 24 calculating a matrix decomposition value from the spatial delay matrix, in which a relation value reflecting a non-linear and non-stationary relationship of the feature values can be obtained by calculating at least one of a DCCA coefficient and a mutual information amount as the relation value of the plurality of types of acoustic feature values, and the disease level of the subject can be predicted on the basis of the relation value.

Abstract: A method for predicting a depressive state using a wearable device. Each data type of biological data for several days from plurality of subjects is converted into data in predetermined time unit (S201). Next, quantiles of the distribution of obtained sample data are determined for each subject and each data type (S202). The standard deviation of distribution of the obtained sample data is calculated for each subject and each data type (S203). The Pearson correlation coefficient is calculated for each combination of data types for each subject (S204). Next, a prediction model for the classification problem of whether subject is in a depressive state is trained by machine leaning (S206), wherein quantiles, standard deviations, and Pearson correlation coefficients extracted from the biological data are features used for an input vector, and an evaluation of existence or non-existence of a depressive state by an expert is a label used as teacher data.

Abstract: A relationship index value computation unit 100A that extracts n words from m texts representing contents of free conversations conducted by m patients whose severity of dementia is known, and computes a relationship index value reflecting a relationship between the m texts and the n words, a prediction model generation unit 14A that generates a prediction model for predicting severity of dementia based on a text index value group including n relationship index values for one text, and a dementia prediction unit 21A that predicts severity of dementia of a patient from a text subjected to prediction by applying the relationship index value computed by the relationship index value computation unit 100A from a text input by a prediction data input unit 20 to a prediction model are included, and severity of dementia can be predicted without performing a mini-mental state examination.