Shanghai Shift Electrics Co., Ltd.

Shanghai Jiao Tong University

Shanghai Sixth People's Hospital

Ling Dai

Zhenwu Xu

Xiaoguo Dai

Bin Sheng

Weiping Jia

Huating Li

A fundus image processing method comprising: receiving a collected fundus image; identifying the fundus image via a first neural network to generate a first feature set of the fundus image; identifying the fundus image via a second neural network to generate a second feature set of the fundus image, wherein the first feature set and the second feature set indicate different lesion attributes of the fundus image; combining the first feature set and the second feature set to obtain a combined feature set of the fundus image; and inputting the combined feature set into a classifier to obtain a classification result.

Fundus image processing method, computer apparatus, and storage medium

Provided is a cleaning device for personal cleaning and care, comprising a cleaning device housing detachably connected to a handle housing, a bracket for supporting the cleaning device, a cleaning element carrier and cleaning elements distributed on the cleaning element carrier, a passive assembly, and a cleaning device transducer. The cleaning device transducer comprises at least two transducer elastic elements; one end of each transducer elastic element is fixedly connected to a corresponding transducer elastic element retainer, and the other end of each transducer elastic element is fixedly connected to an transmission arm of the transducer bracket coupling elastic element; wherein a cleaning device having a natural vibration frequency fis constituted utilizing the bending strain of an elastic material, such that the natural frequency fof the cleaning device is in the range of 100 Hz to 400 Hz, and the relationship between the natural frequency fand the frequency fof the electromagnetic force produced by the drive coil in the handle satisfies: 0.85f&#x3c;f&#x3c;1.05f. The present cleaning device has a compact structure and an aesthetically pleasing shape, rotates steadily, and produces little noise.

Cleaning device

A method of automatically detecting microaneurysm based on multi-sieving convolutional neural network (CNN), includes the following steps of: A), partitioning an image to be detected using random fern and obtaining an auxiliary channel image of the image according to a first partition result; and A), inputting the auxiliary channel image obtained from step A) and the image to a multi-sieving CNN training model to perform a detection and obtaining a microaneurysm detection result of the image. The process of establishing the training model includes: B), using a current microaneurysm diagnostic report as samples and partitioning a lesion image in the microaneurysm diagnostic report using the random fern, and establishing the auxiliary channel image according to a second partition result; B), comparing the obtained auxiliary channel image with a lesion-marked image of pixels, clarifying the samples according to a comparing result and performing the multi-sieving CNN training.

Method of automatically detecting microaneurysm based on multi-sieving convolutional neural network

A cleaning care appliance comprises a transducer comprising a drive shaft and magnets arranged on the left and right sides of the longitudinal axis of the drive shaft. The magnets are independent of each other, the polarity of the magnetic pole of one side of the magnet is opposite to that of the other side in the direction facing driving coil, the angle between the direction of magnetic force lines inside the magnets and the direction of the longitudinal axis of driving coil iron core is greater than 45&#xb0; and smaller than 135&#xb0;, the magnets move with respect to the elastic element retainer. When alternating current flows through the drive coil, the angle between magnet movement direction and the longitudinal axis of the core is greater than 170&#xb0; and smaller than 190&#xb0;. The appliance has low cost, low noise, low damping, and is safe, and reliable.

Personal cleaning care appliance

An adjustable circuit for a personal electric cleaning care appliance comprises comprising a power supply portion and a driver is disclosed. The driver comprises a transducer, a drive coil and an iron core of the drive coil, and a drive shaft of the transducer is equipped with cleaning elements. The adjustable circuit comprises a microchip processor and an H-bridge drive circuit. By connecting a capacitor bank and the drive coil in series at load ends of the H-bridge drive circuit, at least a part of the capacitors of the capacitor bank is controllably connected in series with the drive coil or disconnected from the drive coil, the capacitance of the connected capacitor can be controlled so that the current flowing through the drive coil is controllably limited to a certain value, thereby, the motion amplitude of the cleaning elements can be controlled, and no additional electromagnetic interference will occur.

Adjustable circuit for personal electric cleaning care appliance

A brush head for performing reciprocating rotation comprises a driving wheel and a housing on which a top hole and/or a central through hole are/is distributed in a centering manner. The center line(s) of the holes and a rotation axis (L) of a driving shaft are located at a same line, and the driving shaft penetrates through the central through hole. The fitted region of the driving shaft and the driving wheel enters the through holes distributed on the driving wheel to link the driving shaft and the driving wheel fixedly. A driven part matching a driving part is distributed on a protrusion of the brush disc on which a hollow region is formed. The driving shaft penetrates through or enters the hollow region and limits the brush disc to move. The brush disc is incapable of performing reciprocating rotation when the driving shaft is not installed or is damaged.

Growing community of inventors

Shanghai, China

Ling Dai