![150022-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/150022-Thumbnail%20Image.png?versionId=IfNnmW7ggo44A_uTC2J0e9.t7_kaLeWv&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T025421Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=22889c39672645ca26bd5a42b7e9c623bf33e86d0b6145eb48ce9578c105efd5&itok=Fkfr2vaX)
![147857-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-07/147857-Thumbnail%20Image.png?versionId=I_tKnzjP8Rw1r_m3BbZoF1.YpeOpCSmx&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240530/us-west-2/s3/aws4_request&X-Amz-Date=20240530T154613Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=8e4d369131af0b17c581d5798e8acfa2ce07aa9a9ec537763c095a2dcb701e60&itok=VR0Dr86s)
Mutations in the DNA of somatic cells, resulting from inaccuracies in DNA<br/>replication or exposure to harsh conditions (ionizing radiation, carcinogens), may be<br/>loss-of-function mutations, and the compounding of these mutations can lead to cancer.<br/>Such mutations can come in the form of thymine dimers, N-𝛽 glycosyl bond hydrolysis,<br/>oxidation by hydrogen peroxide or other radicals, and deamination of cytosine to uracil.<br/>However, many cells possess the machinery to counteract the deleterious effects of<br/>such mutations. While eukaryotic DNA repair enzymes decrease the incidence of<br/>mutations from 1 mistake per 10^7 nucleotides to 1 mistake per 10^9 nucleotides, these<br/>mutations, however sparse, are problematic. Of particular interest is a mutation in which<br/>uracil is incorporated into DNA, either by spontaneous deamination of cysteine or<br/>misincorporation. Such mutations occur about one in every 107 cytidine residues in 24<br/>hours. DNA uracil glycosylase (UDG) recognizes these mutations and cleaves the<br/>glycosidic bond, creating an abasic site. However, the rate of this form of DNA repair<br/>varies, depending on the nucleotides that surround the uracil. Most enzyme-DNA<br/>interactions depend on the sequence of DNA (which may change the duplex twist),<br/>even if they only bind to the sugar-phosphate backbone. In the mechanism of uracil<br/>excision, UDG flips the uracil out of the DNA double helix, and this step may be<br/>impaired by base pairs that neighbor the uracil. The deformability of certain regions of<br/>DNA may facilitate this step in the mechanism, causing these regions to be less<br/>mutable. In DNA, base stacking, a form of van der Waals forces between the aromatic<br/>nucleic bases, may make these uracil inclusions more difficult to excise. These regions,<br/>stabilized by base stacking interactions, may be less susceptible to repair by<br/>glycosylases such as UDG, and thus, more prone to mutation.
![148088-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-07/148088-Thumbnail%20Image.png?versionId=dziGSvrvXziiq4weMbUAFWQxDTIKexYW&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240604/us-west-2/s3/aws4_request&X-Amz-Date=20240604T180149Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=da797c14ba9007cf5e63e0ff946af939000e72230b9ad613beec23af2b7329a8&itok=KBLirH05)
Colorimetric assays are an important tool in point-of-care testing that offers several advantages to traditional testing methods such as rapid response times and inexpensive costs. A factor that currently limits the portability and accessibility of these assays are methods that can objectively determine the results of these assays. Current solutions consist of creating a test reader that standardizes the conditions the strip is under before being measured in some way. However, this increases the cost and decreases the portability of these assays. The focus of this study is to create a machine learning algorithm that can objectively determine results of colorimetric assays under varying conditions. To ensure the flexibility of a model to several types of colorimetric assays, three models were trained on the same convolutional neural network with different datasets. The images these models are trained on consist of positive and negative images of ETG, fentanyl, and HPV Antibodies test strips taken under different lighting and background conditions. A fourth model is trained on an image set composed of all three strip types. The results from these models show it is able to predict positive and negative results to a high level of accuracy.
![149856-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/149856-Thumbnail%20Image.png?versionId=WtU2R4ux2OmJmu7zhU56SKCqweNUe8b3&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240530/us-west-2/s3/aws4_request&X-Amz-Date=20240530T153810Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=38e546c43a2d6b889c51604533e8cdd9ff51e00aa7bae2856cf2c95580cc6d8a&itok=1Hec-1pA)
![150218-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/150218-Thumbnail%20Image.png?versionId=GGYJxwG8rmaxkBn5vxZWsbx3fzxfz..e&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T154412Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=3195d2b84af7675a58378bd95092341ac1c42df62aa3545561ee25cf9d82b14c&itok=dOPR_fZ3)
![150143-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/150143-Thumbnail%20Image.png?versionId=vuSeTBiuyRy92PRkieSRkZveAiHj6RZ.&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T172839Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=23b843c9f59ea1079f4d8f01f6dd53e822c0663e565b6ff1302bb1e6c8a08b4f&itok=8s0v2t_S)
![137695-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-05/137695-Thumbnail%20Image.png?versionId=mfggJPdwFn4vLPR_lqYBeK0XzYCw10Qk&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240530/us-west-2/s3/aws4_request&X-Amz-Date=20240530T153837Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=cbe6d3dbfc445874ab790a3bb8f3b6e03aa0d88d721bdf2f3903a2e179d2bd6c&itok=psZSLuo-)
![151753-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/151753-Thumbnail%20Image.png?versionId=XHek65j1Efy4gwasZcOPNH94KDLtZ_z6&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T152040Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=3fc0b41b64de8f5c3591b561ab62386360b226276ef88b1c1088bfd9b95ce86d&itok=rLMFVDj6)
![151652-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-09/151652-Thumbnail%20Image.png?versionId=azerBEMw2SeStXkiX39r.YZqqZusRYLY&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T095125Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=96669a7c5228b04e06d57b9c099710b8d37db7f673f5f362774a26d6ea7b9495&itok=a1X3-BcH)
![150797-Thumbnail Image.png](https://d1rbsgppyrdqq4.cloudfront.net/s3fs-public/styles/width_400/public/2021-08/150797-Thumbnail%20Image.png?versionId=AtWKAdeFDfbipEE5BashJ2H5kzncZHN1&X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIASBVQ3ZQ42ZLA5CUJ/20240619/us-west-2/s3/aws4_request&X-Amz-Date=20240619T170459Z&X-Amz-SignedHeaders=host&X-Amz-Expires=120&X-Amz-Signature=1ced0ed1e86c7a2885f2f4f1560363c0150f19546861489d6e42e4e86e7aec1c&itok=s1fGen83)