Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep learning oral pill identification method based on multiple views and data expansion

A technology of data expansion and deep learning, which is applied in the fields of clinical medicine and nursing, can solve problems that cannot fully meet the needs of drug identification, and achieve the effects of easy transformation, broad application prospects, and improved feature extraction capabilities

Active Publication Date: 2022-07-29
PEOPLES HOSPITAL OF DEYANG CITY +2
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the existing technology can not fully meet the needs of drug identification

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep learning oral pill identification method based on multiple views and data expansion
  • Deep learning oral pill identification method based on multiple views and data expansion
  • Deep learning oral pill identification method based on multiple views and data expansion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0077] lab environment

[0078] The hardware environment and software environment used in this experimental example are shown in Table 1:

[0079] Table 1

[0080]

[0081] The present invention uses the lightweight model MobileNetv2 as the infrastructure. MobileNetv2 was proposed by Google in January 2018. The innovations of the model are the two technologies of Inverted Residuals and Linear Bottlenecks. Designed to improve accuracy and reduce memory usage. The entire model architecture is shown in Table 2

[0082] Table 2

[0083]

[0084] Experimental data

[0085] The experimental data uses 753 pictures of a single pill, and there are a total of 93 categories of pill pictures. High-definition JPG format pictures were collected and shot, and the shooting rules were collected in accordance with the data set collection specifications. The specific quantity and categories are shown in Table 3. table 3

[0086]

[0087]

[0088]

[0089]

[0090] see at...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a deep learning oral pill identification method based on multiple views and data expansion. A database is established by adopting a multi-view and data augmentation method, and a data set is perfected from multiple angles. A lightweight network is used, and a practical model embedded into mobile equipment and small and medium-sized equipment is designed. And combining multiple views with a two-dimensional model, and completing the construction of a practical model after transfer learning. Meanwhile, an incomplete oral pill identification channel is established, and incomplete pills are subjected to template matching to be restored into complete pill pictures and then are identified. The method effectively classifies the medicines with highly similar shapes and colors, assists medical staff in sorting the medicines, and reduces and even avoids life safety problems of patients caused by wrong medicine classification. The overfitting problem caused by small data volume is solved through multi-view database building, data augmentation and transfer learning, a lightweight model MobileNetv2 is adopted as a basic framework, an attention module mechanism is introduced, the parameter quantity of the model is greatly reduced compared with that of a three-dimensional model, and the method is convenient, practical and easy to popularize.

Description

technical field [0001] The invention belongs to the fields of clinical medicine and nursing, and relates to a correct identification method for medicines. Background technique [0002] The hospital shoulders the arduous task of saving lives and helping the wounded. With a lot of work and busy staff, it is unavoidable that some unpackaged medicines are difficult to identify and improperly distributed. [0003] The work of dispensing medicines in hospitals is susceptible to human error. The job of administering medication to patients in settings such as hospitals or nursing wards is that the existing process is a manual process: 1) the correct medication and the correct number of pill sets are put into plastic cups; 2) the pill sets are delivered correctly to the corresponding The patient; 3) the bolus set is administered to the patient at the correct time (eg, no more than 4 hours apart). The process is highly susceptible to human error, and absolute quality assurance is di...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V20/64G06N3/04G06N3/08
CPCG06N3/08G06N3/045
Inventor 向军莲张俊然李南欣谢贤凯刘云飞李杨黄玲唐良友
Owner PEOPLES HOSPITAL OF DEYANG CITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com