BiTTE®-iE Research Use Only Microbial Estimation AI Software

BiTTE®-iE Research Use Only Microbial Estimation AI Software

BiTTE®-iE uses image recognition AI to quickly estimate microorganisms. BiTTE®-iE is also linked to a local antibiogram, i.e. antimicrobial susceptibility testing data.
※ Japanese patent number 7090302.

  • Accuracy

    Approx. 95% for 7-class categorization (GNR/GNC/GPR/GPC/yeast-like fungi/multiple bacteria/no bacteria)

  • Rapid estimation

    Approx. 10 sec. to display the results

  • Link to susceptibility testing data

    Link to local antibiograms

Install the BiTTE®-iE app on your smartphone. Attach the smartphone camera to the microscope eyepiece and capture images.
Select the image of the field of view you want to estimate and tap Start.
You will see the 5-class categorization results in 10 seconds (GNR/GNC/GPR/GPC/yeast-like fungi). Tap the displayed bacteria species (Escherichia coli). ※ Estimated GNR on the left screen
This will display list of antibiotics and their susceptibility to that bacteria species.
Item specifications
Image analysis and display The estimation module analyzes the input image and displays the result of the 7-class categorization with the highest confidence and probability.The confidence is displayed as a percentage from 0 to 1 with two decimal places. The closer the value is to 1, the higher the confidence level.
*These are the 7 morphological classifications of bacteria
1: Gram-positive cocci
2: Gram-positive rods
3: Gram-negative cocci
4: Gram-negative rods
5: Yeast-like fungi
6: Mulitple bacteria
7: No bacteria
Sample Image / Browsing Functions Displays a list of Gram stained images to be determined for each sample.
Input Image validation and confirmation Function Validates and confirms the use of the input image by determining if it contains bacteria.
Multiple Bacteria Determination Function Notifies the user if the input image contains multiple bacteria.

BiTTE®-iE Development History

In 2019, we participated in a clinical needs matching meeting hosted by the Tokyo Metropolitan Organization for Medical Innovation HUB and heard about the clinical needs of the Disease Control and Prevention Center of the National Center for Global Health and Medicine.

The center had a need to develop image AI to estimate the type of bacteria on Gram stained images related to bacterial infections and wanted to use it as an application in anticipation of its use not only in Japan but also in developing countries where drug-resistant bacteria are a problem. Since the image analysis technology had already been established by the parent company, NexGeM Inc., we introduced our capability.

Preliminary analysis was conducted on the sample data, and since relatively high accuracy was achieved, we started a joint research project with the National Center for Global Health and Medicine in April 2020. In October, we also started joint research with Kobe University. I remember how excited and humbled I was to learn from the O'Neill Report and other reports that drug resistance is a major problem worldwide, and to realize that the image analysis technology we have cultivated could make even a small contribution to a problem that has such a large social impact.

We will continue to understand the needs of our customers and make every effort to further improve the accuracy of our products.

General Manager
Isao Miyatsuka

FAQ

A: There is no correction for antimicrobial exposure. As for uneven staining, we provide recommended staining methods, etc. as precautions for use. If you want to check, please contact us through our website.

A: Yes, it works on Android. Analysis results may vary depending on the camera specs including resolution..

A: No. It works online.

A: The accuracy is about 95% for 7-class categorization (GNR/GNC/GPR/GPC/yeast-like fungi/multiple bacteria/no bacteria).

A: Yes, as long as the iPhone can be connected via USB or Wi-Fi.

Inquiry Form

Please fill out the required fields in the inquiry form below to contact us.

I agree to the terms of the Privacy Policy.