Back

Evaluation of the diagnostic value of YiDiXie™-SS in thyroid ultrasound-positive patients

Zhou, H.; Wu, Y.; Li, X.; Sun, C.; Ge, Z.; Chen, W.; Li, Y.; Lin, S.; Zhang, P.; Wang, W.; Chen, S.; Li, W.; Hu, H.; Cui, Y.; He, J.; Lai, Y.

2024-07-05 oncology
10.1101/2024.07.03.24309931 medRxiv
Show abstract

BackgroundThyroid tumor, as an endocrine tumor with increasing incidence, causes a heavy economic burden. Thyroid ultrasound is widely used for screening or diagnosis of thyroid tumors. However, false-positive thyroid ultrasound results can lead to misdiagnosis and wrong puncture biopsy, while false-negative thyroid ultrasound results can lead to missed diagnosis and delayed treatment. There is an urgent need to find a convenient, cost-effective and non-invasive diagnostic method to reduce the false-positive and false-negative rates of thyroid ultrasound. The aim of the present study was to evaluate the diagnostic value of YiDiXie-SS, YiDiXie-HS and YiDiXie-D in thyroid cancer. Patients and methods843 subjects (malignant group, n=820; benign group, n=23) were finally included in this study. The remaining serum samples were collected and tested by YiDiXie all-cancer detection kit. The sensitivity and specificity of YiDiXie-SS were evaluated respectively. ResultsThe sensitivity of YiDiXie-SS was 98.5% (95% CI: 97.5% - 99.2%) and its specificity was 65.2% (95% CI: 44.9% - 81.2%). This means that YiDiXie -SS has an extremely high sensitivity and relatively high specificity in thyroid tumors.YiDiXie-HS has a sensitivity of 84.3% (95% CI: 81.6% - 86.6%) and a specificity of 87.0% (95% CI: 67.9% - 95.5%). This means that YiDiXie-HS has high sensitivity and specificity in thyroid tumors.YiDiXie-D has a sensitivity of 75.5% (95% CI: 72.4% - 78.3%) and a specificity of 95.7% (95% CI: 79.0% - 99.8%). This means that YiDiXie-D has relatively high sensitivity and very high specificity in thyroid tumors.The sensitivity of YiDiXie-SS in patients with positive thyroid ultrasound was 98.4% (95% CI: 97.3% - 99.1%) and the specificity was 64.3% (95% CI: 38.8% - 83.7%). This means that the application of YiDiXie -SS reduced the false-positive thyroid ultrasound rate by 64.3% (95% CI: 38.8% - 83.7%) with essentially no increase in malignant tumor underdiagnosis.The sensitivity of YiDiXie-HS in thyroid ultrasound-negative patients was 90.0% (95% CI: 79.9% - 95.3%) and the specificity 88.9% (95% CI: 56.9%). (95% CI: 56.5% - 99.4%). This means that the application of YiDiXie-HS reduced the false negative rate of thyroid ultrasound by 90.0% (95% CI: 79.9% - 95.3%).YiDiXie -D has a sensitivity of 75.3% (95% CI: 72.1% - 78.2%) and a specificity of 92.9% (95% CI: 68.5% - 99.6%) in patients with positive thyroid ultrasound. This means that YiDiXie-D reduces the rate of false-positive thyroid ultrasounds by 92.9% (95% CI: 68.5% - 99.6%). YiDiXie-D has a sensitivity of 78.3% (95% CI: 66.4% - 86.9%) and a specificity of 100% (95% CI: 70.1% - 100%) in patients with negative thyroid ultrasounds. This means that YiDiXie-D reduces the false-negative rate of thyroid ultrasound by 78.3% (95% CI: 66.4% - 86.9%) while maintaining high specificity. ConclusionYiDiXie-SS has extremely high sensitivity and relatively high specificity in thyroid tumors.YiDiXie-HS has high sensitivity and high specificity in thyroid tumors.YiDiXie -D has relatively high sensitivity and extremely high specificity in thyroid tumors. YiDiXie-SS significantly reduces thyroid ultrasound false-positive rates with essentially no increase in delayed treatment for thyroid cancer.YiDiXie -HS significantly reduces thyroid ultrasound false-negative rates.YiDiXie-D significantly reduces thyroid ultrasound false-positive rates or significantly reduces its false-negative rates while maintaining high specificity. YiDiXie tests have vital diagnostic value in thyroid cancer, and are expected to solve the problems of "high false-positive rate" and "high false-negative rate" of thyroid ultrasound. Clinical trial numberChiCTR2200066840.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.