Skip to content
Snippets Groups Projects
Commit 5b12856f authored by Siyuan Lyu's avatar Siyuan Lyu
Browse files

Update 2 files

- /wiki/pages/hardware.html
- /wiki/pages/software.html
parent 249147cb
No related branches found
No related tags found
No related merge requests found
Pipeline #491702 passed
......@@ -74,7 +74,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/1.svg">
</div>
<div class="disc">Figure 1. An overview of the microfluidic chip system.</div>
<div class="disc">Figure 1. Overview of the microfluidic ship system</div>
</div>
</div>
......@@ -108,6 +108,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/pic2.svg">
</div>
<div class="disc">Figure 2. The microfluidic chip structure.</div>
</div>
<h2 id="2-1">&bull;&emsp;Accessory plumbing system</h2>
......@@ -118,9 +119,10 @@
</P>
<div class="img-wrap">
<div class="border">
<div class="border" style="width: 50%;">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/pic3.svg">
</div>
<div class="disc">Figure 3. The structure of our accessory plumbing system.</div>
</div>
<h2 id="2-2">&bull;&emsp;Jig of the chip</h2>
......@@ -144,6 +146,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/pic4.svg">
</div>
<div class="disc">Figure 4. The jig of the chip.</div>
</div>
</div>
......@@ -155,6 +158,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/pic5.svg">
</div>
<div class="disc">Figure 5. The cost analysis of the hardware.</div>
</div>
</div>
......@@ -199,6 +203,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/pic6.svg">
</div>
<div class="disc">Figure 6. The 0.17mm coverslip enable us to test cell under the 60x microscope.</div>
</div>
<p>2. Design the Jig of the chip to fix the microfluidic chip and reduce the stress on the PDMS material from stainless-steel tube. Thus, the focus shift was largely reduced.
</p>
......@@ -206,6 +211,7 @@
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/hardware/11.jpg">
</div>
<div class="disc">Figure 7. The focus shift was largely reduced by using the chip jig.</div>
</div>
......
......@@ -104,7 +104,7 @@
<h2 id="2-1">&bull;&emsp;Dot matrix segmentation</h2>
<div class="img-wrap">
<div class="border">
<img src="https://static.igem.wiki/teams/5371/dry-lab/software/2.png">
<img src="https://static.igem.wiki/teams/5371/dry-lab/software/2.png" style="transform: scale(1.4);transform-origin: left;margin-bottom: -9vh;">
</div>
<div class="disc">Figure 2. The YOLOv8 structure for dot matrix segmentation.</div>
</div>
......@@ -115,7 +115,7 @@
<p>The process of manually annotating the massive amount of image data obtained from microscopy is not only time-consuming but also labor-intensive. To address this, we have implemented machine learning models for the automatic recognition of binary matrix values, streamlining the batch processing workflow. Focusing on the binary nature of these matrices, we simplified the model by utilizing only the grayscale channel and applied histogram equalization to boost contrast. We utilized CNNs, renowned for their prowess in image processing, with a compact architecture consisting of two convolutional layers succeeded by two fully connected layers (Fig 3). Compared and tested on different models, the CNN is sufficient to meet our needs. To enhance the robustness of the model and reduce overfitting, we expanded the dataset by generating additional images through transformations based on the manually annotated samples. This approach significantly improves the efficiency of biological image processing using machine learning techniques with good model performance (check performance for details).
</p>
<div class="img-wrap">
<div class="border">
<div class="border" style="width: 50%;">
<img src="https://static.igem.wiki/teams/5371/dry-lab/software/3.png">
</div>
<div class="disc">Fig 3. The CNN structure for dot recognition.</div>
......@@ -128,7 +128,7 @@
<p>Based on the latest iteration of the cellpose model, cyto3, we have trained our model to precisely recognize and segmented yeast cells [3]. In our implementation with tagged images with its number, the model is initiated with image input, and proceeded with denoising, and calibrates cell diameter pixels to optimize recognition accuracy. Then we manually modified incorrectly segmented yeast cells, thereby ensuring the precise segmentation of yeast (Fig 4).
</p>
<div class="img-wrap">
<div class="border">
<div class="border" style="width: 50%;">
<img src="https://static.igem.wiki/teams/5371/dry-lab/software/4.png">
</div>
<div class="disc">Figure 4. The results of trained cyto3 model recognition. The colored dots represent an identified yeast.</div>
......
0% Loading or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment