A CT-Based Deep Learning Radiomics Nomogram to Predict Histological Grades of Head and Neck Squamous Cell Carcinoma

Zheng, Ying-Mei; Che, Junyi; Yuan, Ming-Gang; Wu, Zhen; Pang, Jing; Zhou, Ruizhi; Li, Xiaoli; Dong, Cheng

doi:10.1016/j.acra.2022.11.007

Cited by 19 publications

(8 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among these models, the Rad + ResNet34_SR combination model emerged as the top performer, achieving an Accuracy of 0.929, AUC of 0.952, Sensitivity of 0.909, and Speci city of 0.941. This aligns with the ndings of Jun Zhang et al, demonstrating the superior e cacy of the joint model that leverages deep learning radiomics for clinical applications 34,35 . This model's high accuracy and detection rate underscore its signi cant potential to enhance clinical diagnosis and treatment.…”

Section: Discussionsupporting

confidence: 88%

Deep Transfer Learning-based 3D super-resolution Radiomics modeling of CT to discriminate Brucella Spondylitis from Tuberculous Spondylitis: A multicenter, retrospective study

Wang,

Qi,

Zhang

et al. 2024

Preprint

View full text Add to dashboard Cite

To improve the accuracy in distinguishing between Tuberculosis Spondylitis (TBS) and Brucella spondylitis (BS) through the creation of radiomics deep Learning models. These models incorporate Deep Learning techniques and leverage CT image enhanced by Generative Adversarial Network-based Super-Resolution (GAN-SR).94 patients from Ordos Central and Second People's Hospitals diagnosed with BS or TBS. These patients were divided into a training set (n = 65) and a validation set (n = 29). We utilized conventional CT images to generate SR-CT with twice SR based on GANs-SR. The ROIs underwent manual segmentation, and extraction of Radiomics features, and ResNet18 and ResNet34 were used as Deep Learning models for training and DL feature extraction. Feature fusion and selected were performed, and four models were built based on MLP: clinical; radiomics (Rad); Deep Learning (DL) and combined models. The model efficacy was evaluated using ROC and DCA, etc. In terms of efficacy, the integrated model outperformed others, showing the highest AUC, with the DL model following closely behind. Notably, ResNet34 demonstrated superior performance over ResNet18, SR enhanced model performance across metrics for SR-CT compared to CT. The combination of Rad + ResNet34_SR achieved the best results, with AUC of 0.952, Sensitivity of 0.909, and Specificity of 0.941. DCA curve confirmed the model's clinical decision-making potential.In conclusion, our study creates a 3D Super-Resolution model based on SR-CT images using Deep Learning Radiomics to differentiate between TBS and BS, which is significantly enhanced compared to CT images.

show abstract

Section: Discussionsupporting

confidence: 88%

Deep Transfer Learning-based 3D super-resolution Radiomics modeling of CT to discriminate Brucella Spondylitis from Tuberculous Spondylitis: A multicenter, retrospective study

Wang,

Qi,

Zhang

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…This model demonstrated its ability to improve the accuracy and reliability of predictions significantly. Numerous studies have consistently shown that DLR outperforms Rad_Sig and DTL_Sig when used individually [ 38 – 41 ]. The fusion of radiomics and DL in this model involved two distinct methods: feature and result fusion.…”

Section: Resultsmentioning

confidence: 99%

Development and validation of an ultrasound-based deep learning radiomics nomogram for predicting the malignant risk of ovarian tumours

Du,

Xiao,

Guo

et al. 2024

BioMed Eng OnLine

View full text Add to dashboard Cite

Background The timely identification and management of ovarian cancer are critical determinants of patient prognosis. In this study, we developed and validated a deep learning radiomics nomogram (DLR_Nomogram) based on ultrasound (US) imaging to accurately predict the malignant risk of ovarian tumours and compared the diagnostic performance of the DLR_Nomogram to that of the ovarian-adnexal reporting and data system (O-RADS). Methods This study encompasses two research tasks. Patients were randomly divided into training and testing sets in an 8:2 ratio for both tasks. In task 1, we assessed the malignancy risk of 849 patients with ovarian tumours. In task 2, we evaluated the malignancy risk of 391 patients with O-RADS 4 and O-RADS 5 ovarian neoplasms. Three models were developed and validated to predict the risk of malignancy in ovarian tumours. The predicted outcomes of the models for each sample were merged to form a new feature set that was utilised as an input for the logistic regression (LR) model for constructing a combined model, visualised as the DLR_Nomogram. Then, the diagnostic performance of these models was evaluated by the receiver operating characteristic curve (ROC). Results The DLR_Nomogram demonstrated superior predictive performance in predicting the malignant risk of ovarian tumours, as evidenced by area under the ROC curve (AUC) values of 0.985 and 0.928 for the training and testing sets of task 1, respectively. The AUC value of its testing set was lower than that of the O-RADS; however, the difference was not statistically significant. The DLR_Nomogram exhibited the highest AUC values of 0.955 and 0.869 in the training and testing sets of task 2, respectively. The DLR_Nomogram showed satisfactory fitting performance for both tasks in Hosmer–Lemeshow testing. Decision curve analysis demonstrated that the DLR_Nomogram yielded greater net clinical benefits for predicting malignant ovarian tumours within a specific range of threshold values. Conclusions The US-based DLR_Nomogram has shown the capability to accurately predict the malignant risk of ovarian tumours, exhibiting a predictive efficacy comparable to that of O-RADS.

show abstract

“…Deep learning (DL) neural networks exemplify the successful integration of artificial intelligence's automated processing into clinical practice, showcasing outstanding performance in tasks like image processing and classification, including applications in ultrasound and CT (computerised tomography) 26 - 28 . Research indicates that the classification ability of deep neural networks for dermoscopic images can rival that of dermatologists 29 .…”

Section: Introductionmentioning

confidence: 99%

Dermoscopy-based Radiomics Help Distinguish Basal Cell Carcinoma and Actinic Keratosis: A Large-scale Real-world Study Based on a 207-combination Machine Learning Computational Framework

Guan,

Yuan,

et al. 2024

J. Cancer

View full text Add to dashboard Cite

This study has used machine learning algorithms to develop a predictive model for differentiating between dermoscopic images of basal cell carcinoma (BCC) and actinic keratosis (AK). We compiled a total of 904 dermoscopic images from two sources -the public dataset (HAM10000) and our proprietary dataset from the First Affiliated Hospital of Dalian Medical University (DAYISET 1) -and subsequently categorised these images into four distinct cohorts. The study developed a deep learning model for quantitative analysis of image features and integrated 15 machine learning algorithms, generating 207 algorithmic combinations through random combinations and cross-validation. The final predictive model, formed by integrating XGBoost with Lasso regression, exhibited effective performance in the differential diagnosis of BCC and AK. The model demonstrated high sensitivity in the training set and maintained stable performance in three validation sets. The area under the curve (AUC) value reached 1.000 in the training set and an average of 0.695 in the validation sets. The study concludes that the constructed discriminative diagnostic model based on machine learning algorithms has excellent predictive capabilities that could enhance clinical decision-making efficiency, reduce unnecessary biopsies, and provide valuable guidance for further treatment.

show abstract

A CT-Based Deep Learning Radiomics Nomogram to Predict Histological Grades of Head and Neck Squamous Cell Carcinoma

Cited by 19 publications

References 30 publications

Deep Transfer Learning-based 3D super-resolution Radiomics modeling of CT to discriminate Brucella Spondylitis from Tuberculous Spondylitis: A multicenter, retrospective study

Deep Transfer Learning-based 3D super-resolution Radiomics modeling of CT to discriminate Brucella Spondylitis from Tuberculous Spondylitis: A multicenter, retrospective study

Development and validation of an ultrasound-based deep learning radiomics nomogram for predicting the malignant risk of ovarian tumours

Dermoscopy-based Radiomics Help Distinguish Basal Cell Carcinoma and Actinic Keratosis: A Large-scale Real-world Study Based on a 207-combination Machine Learning Computational Framework

Contact Info

Product

Resources

About