Nucleic acid amplification detection methods and portable nucleic acid amplification detection devices

Abstract

A nucleic acid amplification detection method is applicable to determining whether nucleic acid amplification of at least one nucleic acid reaction sample is valid. The method is performed by a device and includes the following steps: (A) capturing an image of the at least one nucleic acid reaction sample to generate a color image; (B) extracting pixel component values ​​of at least one preset region of interest in the color image to generate at least one interpretation index value; (C) generating at least one first interpretation result based on the at least one interpretation index value and a preset threshold; and (D) identifying the at least one nucleic acid reaction sample in the color image using a pre-trained deep learning object detection model to generate at least one second interpretation result.

Claims

1. A nucleic acid amplification detection method, applicable to determining whether nucleic acid amplification of at least one nucleic acid reaction sample is valid, executed by a device storing a pre-trained deep learning object detection model, the nucleic acid amplification detection method comprising the following steps: (A) capturing an image of the at least one nucleic acid reaction sample to generate a color image; (B) extracting pixel component values ​​of at least one preset region of interest corresponding to the at least one nucleic acid reaction sample from the color image to generate at least one interpretation index value corresponding to the at least one nucleic acid reaction sample; (C) Based on the at least one interpretation index value and a preset threshold, generate at least one first interpretation result corresponding to the at least one nucleic acid reaction sample, each first interpretation result indicating whether the nucleic acid amplification of the corresponding at least one nucleic acid reaction sample is established; and (D) Based on the color image, use the deep learning object detection model to identify the at least one nucleic acid reaction sample in the color image to generate at least one second interpretation result corresponding to the at least one nucleic acid reaction sample, each second interpretation result indicating whether the nucleic acid amplification of the corresponding at least one nucleic acid reaction sample is established.

2. The nucleic acid amplification detection method as described in claim 1 further includes the following step between step (A) and step (B): (E) performing brightness normalization processing on the color image; wherein, In step (B), the pixel component values ​​of the at least one preset region of interest are extracted based on the brightness-normalized color image.

3. The nucleic acid amplification detection method as described in claim 1, wherein, Step (B) includes the following sub-steps: (B-1) For each preset region of interest in the color image, extract multiple red pixel component values, multiple green pixel component values, and multiple blue pixel component values ​​of all pixels; (B-2) For each preset region of interest in the color image, calculate an average value of the red pixel component, an average value of the green pixel component, and an average value of the blue pixel component; and (B-3) For each preset region of interest in the color image, use the ratio of two of the average values ​​of the red pixel component, the average values ​​of the green pixel component, and the average values ​​of the blue pixel component as an interpretation index value.

4. The nucleic acid amplification detection method as described in claim 3, wherein, In sub-step (B-3), the interpretation index value of the nucleic acid reaction sample is the average value of the green pixel component divided by the average value of the blue pixel component.

5. The nucleic acid amplification detection method as described in claim 1, wherein the device is communicatively connected to a remote device, and after step (D) further includes the following steps: (F) transmitting the pixel component values ​​of the at least one preset region of interest, the first interpretation result, and the second interpretation result to the remote device.

6. A portable nucleic acid amplification detection device, suitable for determining whether nucleic acid amplification of at least one nucleic acid reaction sample is successful, the portable nucleic acid amplification detection device comprising: a storage unit storing a pre-trained deep learning object detection model; an imaging unit; and a processing unit electrically connected to the storage unit and the imaging unit; wherein, The processing unit controls the imaging unit to capture the at least one nucleic acid reaction sample to generate a color image. Based on at least one preset region of interest (ROI) in the color image corresponding to the at least one nucleic acid reaction sample, the processing unit extracts pixel component values ​​of the preset ROI to generate at least one interpretation index value corresponding to the at least one nucleic acid reaction sample. Based on the at least one interpretation index value and a preset threshold, at least one first interpretation result corresponding to the at least one nucleic acid reaction sample is generated. Each first interpretation result indicates whether nucleic acid amplification of the corresponding at least one nucleic acid reaction sample is successful. Based on the color image, the processing unit uses the deep learning object detection model stored in the storage unit to identify the at least one nucleic acid reaction sample in the color image to generate at least one second interpretation result corresponding to the at least one nucleic acid reaction sample. Each second interpretation result indicates whether nucleic acid amplification of the corresponding at least one nucleic acid reaction sample is successful.

7. The portable nucleic acid amplification and detection device as described in claim 6, wherein, The processing unit performs luminance normalization on the color image, and extracts the pixel component values ​​of the at least one preset region of interest based on the luminance normalized color image.

8. The portable nucleic acid amplification and detection device as described in claim 6, wherein, For each preset region of interest in the color image, the processing unit extracts multiple red pixel component values, multiple green pixel component values, and multiple blue pixel component values ​​of all pixels, and calculates an average value of the red pixel component, an average value of the green pixel component, and an average value of the blue pixel component. The ratio of any two of the average values ​​of the red pixel component, the green pixel component, and the blue pixel component is used as a judgment index value.

9. The portable nucleic acid amplification and detection device as described in claim 8, wherein, The interpretation index for this nucleic acid reaction sample is the average value of the green pixel component divided by the average value of the blue pixel component.

10. The portable nucleic acid amplification and detection device as described in claim 6 further comprises: a communication unit, communicatively connected to a remote device; wherein, The processing unit transmits the pixel component values ​​of the at least one preset region of interest, the first judgment result, and the second judgment result to the remote device via the communication unit.

Images