Lymph node metastasis prediction of papillary thyroid carcinoma based on transfer learning radiomics

SUITOR: Selecting the number of mutational signatures through cross-validation

For de novo mutational signature analysis, the critical first step is to decide how many signatures should be expected in a cancer genomics study. An incorrect number could mislead downstream analyses. Here we present SUITOR (Selecting the nUmber of mutatIonal signaTures thrOugh cRoss-validation), an unsupervised cross-validation method that requires little assumptions and no numerical approximations to select the optimal number of signatures without overfitting the data. In vitro studies and in silico simulations demonstrated that SUITOR can correctly identify signatures, some of which were missed by other widely used methods. Applied to 2,540 whole-genome sequenced tumors across 22 cancer types, SUITOR selected signatures with the smallest prediction errors and almost all signatures of breast cancer selected by SUITOR were validated in an independent breast cancer study. SUITOR is a powerful tool to select the optimal number of mutational signatures, facilitating downstream analyses with etiological or therapeutic importance.

The anti-cancer role of microRNA-143 in papillary thyroid carcinoma by targeting high mobility group AT-hook 2

Papillary thyroid carcinoma (PTC), a common thyroid cancer (TC) subtype, rapidly increases in occurrence. MicroRNAs (miRNAs), which are non-coding small RNAs, have been demonstrated to play a role in cancer pathogenic mechanisms. Although miR-143 is involved in suppressing certain malignant tumor progression, its biological role is unknown in PTC. The present study found that miR-143 levels were strongly lower in PTC patient samples and cell lines, implying that miR-143 may play a biological role in PTC. Down-regulation of miR-143 resulted in the increased expression of HMGA2. Furthermore, HMGA2 was found to be a direct target of miR-143. A dual-luciferase assay confirmed a direct binding site for miR-143 was confirmed on HMGA2 using a dual-luciferase assay. Next, over-expression of miR-143 suppressed PTC cell growth as analyzed by MTT, clone formation, and Ki-67 immunofluorescence staining assays. miR-143 mimics transfection downregulated the expression of PCNA, CDK4, CDK1, and Cyclin E1. In addition, wound healing and trans-well assays revealed that miR-143 up-regulation inhibited PTC cells invasion and migration. Co-transfection of HMGA2 expression vector restored HMGA2 expression and rescued PTC cells proliferation capability in miR-143 mimics transfected PTC cells, indicating that miR-143 inhibited PTC cells proliferation via HMGA2. These observations were also obtained in xenografts experiments in nude mice. Altogether, our study shed light on miR-143ʹs anti-cancer biological functions in PTC progression through targeting HMGA2, suggesting that restoration of miR-143 could be a potential therapeutic approach for PTC treatment.

Deep multimodal learning for lymph node metastasis prediction of primary thyroid cancer

Objective. The incidence of primary thyroid cancer has risen steadily over the past decades because of overdiagnosis and overtreatment through the improvement in imaging techniques for screening, especially in ultrasound examination. Metastatic status of lymph nodes is important for staging the type of primary thyroid cancer. Deep learning algorithms based on ultrasound images were thus developed to assist radiologists on the diagnosis of lymph node metastasis. The objective of this study is to integrate more clinical context (e.g., health records and various image modalities) into, and explore more interpretable patterns discovered by, deep learning algorithms for the prediction of lymph node metastasis in primary thyroid cancer patients.Approach. A deep multimodal learning network was developed in this study with a novel index proposed to compare the contribution of different modalities when making the predictions.Main results. The proposed multimodal network achieved an average F1 score of 0.888 and an average area under the receiver operating characteristic curve (AUC) value of 0.973 in two independent validation sets, and the performance was significantly better than that of three single-modality deep learning networks. Moreover, among three modalities used in this study, the deep multimodal learning network relied generally more on image modalities than the data modality of clinic records when making the predictions.Significance. Our work is beneficial to prospective clinic trials of radiologists on the diagnosis of lymph node metastasis in primary thyroid cancer, and will better help them understand how the predictions are made in deep multimodal learning algorithms.

Role of Radiomics-Based Baseline PET/CT Imaging in Lymphoma: Diagnosis, Prognosis, and Response Assessment

Radiomic analysis provides information on the underlying tumour heterogeneity in lymphoma, reflecting the real-time evolution of malignancy. 2-Deoxy-2-[¹⁸F] fluoro-D-glucose positron emission tomography ([¹⁸F] FDG PET/CT) imaging is recommended before, during, and at the end of treatment for almost all lymphoma patients. This methodology offers high specificity and sensitivity, which can aid in accurate staging and assist in prompt treatment. Pretreatment [¹⁸F] FDG PET/CT-based radiomics facilitates improved diagnostic ability, guides individual treatment regimens, and boosts outcome prognosis based on heterogeneity as well as the biological, pathological, and metabolic status of the lymphoma. This technique has attracted considerable attention given its numerous applications in medicine. In the current review, we will briefly describe the basic radiomics workflow and types of radiomic features. Details of current applications of baseline [¹⁸F] FDG PET/CT-based radiomics in lymphoma will be discussed, such as differential diagnosis from other primary malignancies, diagnosis of bone marrow involvement, and response and prognostic prediction. We will also describe how this technique provides a unique noninvasive platform to assess tumour heterogeneity. Newly emerging PET radiotracers and multimodality technology will improve diagnostic specificity and further clarify tumor biology and even genetic variations in lymphoma, potentially promoting the development of precision medicine.

Lack of Association Between Helicobacter pylori Infection and the Risk of Thyroid Nodule Types: A Multicenter Case-Control Studyin China

The prevalence of Helicobacter pylori infection is high worldwide, while numerous research has focused on unraveling the relationship between H. pylori infection and extragastric diseases. Although H. pylori infection has been associated with thyroid diseases, including thyroid nodule (TN), the relationship has mainly focused on potential physiological mechanisms and has not been validated by large population epidemiological investigations. Therefore, we thus designed a case-control study comprising participants who received regular health examination between 2017 and 2019. The cases and controls were diagnosed via ultrasound, while TN types were classified according to the guidelines of the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS). Moreover, H. pylori infection was determined by C14 urea breath test, while its relationship with TN type risk and severity was analyzed using binary and ordinal logistic regression analyses. A total of 43,411 participants, including 13,036 TN patients and 30,375 controls, were finally recruited in the study. The crude odds ratio (OR) was 1.07 in Model 1 (95% CI = 1.03-1.14) without adjustment compared to the H. pylori non-infection group. However, it was negative in Model 2 (OR = 1.02, 95% CI = 0.97-1.06) after being adjusted for gender, age, body mass index (BMI), and blood pressure and in Model 3 (OR = 1.01, 95% CI = 0.97-1.06) after being adjusted for total cholesterol, triglyceride, low-density lipoprotein, and high-density lipoprotein on the basis of Model 2. Control variables, including gender, age, BMI, and diastolic pressure, were significantly correlated with the risk of TN types. Additionally, ordinal logistic regression results revealed that H. pylori infection was positively correlated with malignant differentiation of TN (Model 1: OR = 1.06, 95% CI = 1.02-1.11), while Model 2 and Model 3 showed negative results (Model 2: OR = 1.01, 95% CI = 0.96-1.06; Model 3: OR = 1.01, 95% CI = 0.96-1.05). In conclusion, H. pylori infection was not significantly associated with both TN type risk and severity of its malignant differentiation. These findings provide relevant insights for correcting possible misconceptions regarding TN type pathogenesis and will help guide optimization of therapeutic strategies for thyroid diseases.

Role of color-coded virtual touch tissue imaging in suspected thyroid nodules

BACKGROUND

Conventional ultrasound (US) is the most widely used imaging test for thyroid nodule surveillance.

OBJECTIVE

We used the color-coded virtual touch tissue imaging (VTI) in the Acoustic Radiation Force Impulse (ARFI) technique to assess the hardness of the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) TR3-5 nodules. The ability of color-coded VTI (CV) to discriminate between benign and malignant nodules was investigated.

METHODS

In this retrospective study, US and CV were performed on 211 TR3-5 thyroid lesions in 181 consecutive patients. All nodules were operated on to obtain pathological results. A multivariate logistic regression model was chosen to integrate the data obtained from the US and CV.

RESULTS

The area under the receiver operating characteristic (ROC) curve for the model was 0.945 (95% CI, 0.914 to 0.976). The cutoff value of predictive probability for diagnosing malignant thyroid nodules was 10.64%, the sensitivity was 94.43%, and the specificity was 83.12%. Through comparing with US and CV, respectively, it had been observed that the regression model had the best performance (all P< 0.001). However, when the US was compared with CV, the difference was not significant (P= 0.3304).

CONCLUSIONS

A combination of US and CV should be recommended for suspected malignant thyroid nodules in clinical practice.

A radiomics nomogram for the ultrasound-based evaluation of central cervical lymph node metastasis in papillary thyroid carcinoma

PURPOSE

To develop and validate a radiomics nomogram based on ultrasound (US) to predict central cervical lymph node (LN) metastasis in patients with papillary thyroid carcinoma (PTC).

METHODS

PTC patients with pathologically confirmed presence or absence of central cervical LN metastasis in our hospital between March 2021 and November 2021 were enrolled as the training cohort. Radiomics features were extracted from the preoperative US images, and a radiomics signature was constructed. Univariate and multivariate logistic regression analyses were used to screen out the independent risk factors, and a radiomics nomogram was established. The performance of the model was verified in the independent test cohort of PTC patients who underwent thyroidectomy and cervical LN dissection in our hospital from December 2021 to March 2022.

RESULTS

In the independent test cohort, the radiomics model based on long-axis cross-section and short-axis cross-section images outperformed the radiomics models based on either one of these sections (the area under the curve (AUC), 0.69 vs. 0.62 and 0.66). The radiomics signature consisted of 4 selected features. The US radiomics nomogram included the radiomics signature, age, gender, BRAF V600E mutation status, and extrathyroidal extension (ETE) status. In the independent test cohort, the AUC of the receiver operating curve(ROC) of this nomogram was 0.76, outperformingthe clinical model and the radiomics model (0.63 and 0.69, respectively), and also much better than preoperative US examination (AUC, 0.60). Decision curve analysis indicated that the radiomics nomogram was clinically useful.

CONCLUSIONS

This study presents an efficient and useful US radiomics nomogram that can provide comprehensive information to assist clinicians in the individualized preoperative prediction of central cervical LN metastasis in PTC patients.

Recurrence Risk of Liver Cancer Post-hepatectomy Using Machine Learning and Study of Correlation With Immune Infiltration

Postoperative recurrence of liver cancer is the main obstacle to improving the survival rate of patients with liver cancer. We established an mRNA-based model to predict the risk of recurrence after hepatectomy for liver cancer and explored the relationship between immune infiltration and the risk of recurrence after hepatectomy for liver cancer. We performed a series of bioinformatics analyses on the gene expression profiles of patients with liver cancer, and selected 18 mRNAs as biomarkers for predicting the risk of recurrence of liver cancer using a machine learning method. At the same time, we evaluated the immune infiltration of the samples and conducted a joint analysis of the recurrence risk of liver cancer and found that B cell, B cell naive, T cell CD4⁺ memory resting, and T cell CD4⁺ were significantly correlated with the risk of postoperative recurrence of liver cancer. These results are helpful for early detection, intervention, and the individualized treatment of patients with liver cancer after surgical resection, and help to reveal the potential mechanism of liver cancer recurrence.

MRI-based peritumoral radiomics analysis for preoperative prediction of lymph node metastasis in early-stage cervical cancer: A multi-center study

PURPOSE

To evaluate intra- and preitumoral radiomics on the contrast-enhanced T1-weighted (CE-T1) and T2-weighted (T2W) MRI for predicting the LNM, and develop a nomogram for potential clinical uses.

METHODS

We enrolled 169 cervical cancer cases who underwent CE-T1 and T2W MR scans from two hospitals between Dec. 2015 and Sep. 2021. Intra- and peritumoral features were extracted separately and selected by the least absolute shrinkage and selection operator (LASSO) regression. Radiomics signatures were built using the selected features from different regions. Clinical parameters were evaluated by statistical analysis. The nomogram was developed combining the multi-regional radiomics signature and the most predictive clinical parameters.

RESULTS

Five radiomics features were finally selected from the peritumoral regions with 1 and 3 mm distances in the CE-T1 and T2W MRI, respectively. The nomogram incorporating multi-regional combined radiomics signature, MR-reported LN status and tumor diameter achieved the highest AUCs in the training (nomogram vs. combined radiomics signature vs. clinical model, 0.891 vs. 0.830 vs. 0.812), internal validation (nomogram vs. combined radiomics signature vs. clinical model, 0.863 vs. 0.853 vs. 0.816) and external validation (nomogram vs. combined radiomics signature vs. clinical model, 0.804 vs. 0.701 vs. 0.787) cohort. DCA suggested good clinical usefulness of our developed models.

CONCLUSION

The current work suggested clinical potential for intra- and peritumoral radiomics with multi-modal MRI for preoperative predicting LNM.

Deep learning radiomics model related with genomics phenotypes for lymph node metastasis prediction in colorectal cancer

BACKGROUND AND PURPOSE

The preoperative lymph node (LN) status is important for the treatment of colorectal cancer (CRC). Here, we established and validated a deep learning (DPL) model for predicting lymph node metastasis (LNM) in CRC.

MATERIALS AND METHODS

A total of 423 CRC patients were divided into cohort 1 (training set, n = 238, testing set, n = 101) and cohort 2 (validation set, n = 84). Among them, 84 patients' tumour tissues were collected for RNA sequencing. The DPL features were extracted from enhanced venous-phase computed tomography of CRC using an autoencoder. A DPL model was constructed with the least absolute shrinkage and selection operator algorithm. Carcinoembryonic antigen and carbohydrate antigen 19-9 were incorporated into the DPL model to construct a combined model. The model performance was assessed by receiver operating characteristic curves, calibration curves and decision curves. The correlations between DPL features, which have been selected, and genes were analysed by Spearman correlation, and the genes correlated with DPL features were used to transcriptomic analysis.

RESULTS

The DPL model, integrated with 20 DPL features, showed a good discrimination performance in predicting the LNM, with areas under the curves (AUCs) of 0.79, 0.73 and 0.70 in the training set, testing set and validation set, respectively. The combined model had a better performance, with AUCs of 0.81, 0.77 and 0.73 in the three sets, respectively. Decision curve analysis confirmed the clinical application value of the DPL model and combined model. Furthermore, catabolic processes and immune-related pathways were identified and related with the selected DPL features.

CONCLUSION

This study presented a DPL model and a combined model for LNM prediction. We explored the potential genomic phenotypes related with DPL features. In addition, the model could potentially be utilized to facilitate the individualized prediction of LNM in CRC.

Differentiation of Brain Abscess From Cystic Glioma Using Conventional MRI Based on Deep Transfer Learning Features and Hand-Crafted Radiomics Features

Objectives: To develop and validate the model for distinguishing brain abscess from cystic glioma by combining deep transfer learning (DTL) features and hand-crafted radiomics (HCR) features in conventional T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI).

Methods: This single-center retrospective analysis involved 188 patients with pathologically proven brain abscess (102) or cystic glioma (86). One thousand DTL and 105 HCR features were extracted from the T1WI and T2WI of the patients. Three feature selection methods and four classifiers, such as k-nearest neighbors (KNN), random forest classifier (RFC), logistic regression (LR), and support vector machine (SVM), for distinguishing brain abscess from cystic glioma were compared. The best feature combination and classifier were chosen according to the quantitative metrics including area under the curve (AUC), Youden Index, and accuracy.

Results: In most cases, deep learning-based radiomics (DLR) features, i.e., DTL features combined with HCR features, contributed to a higher accuracy than HCR and DTL features alone for distinguishing brain abscesses from cystic gliomas. The AUC values of the model established, based on the DLR features in T2WI, were 0.86 (95% CI: 0.81, 0.91) in the training cohort and 0.85 (95% CI: 0.75, 0.95) in the test cohort, respectively.

Conclusions: The model established with the DLR features can distinguish brain abscess from cystic glioma efficiently, providing a useful, inexpensive, convenient, and non-invasive method for differential diagnosis. This is the first time that conventional MRI radiomics is applied to identify these diseases. Also, the combination of HCR and DTL features can lead to get impressive performance.

Classification and Diagnosis of Residual Thyroid Tissue in SPECT Images Based on Fine-Tuning Deep Convolutional Neural Network

Patients with thyroid cancer will take a small dose of 131I after undergoing a total thyroidectomy. Single-photon emission computed tomography (SPECT) is used to diagnose whether thyroid tissue remains in the body. However, it is difficult for human eyes to observe the specificity of SPECT images in different categories, and it is difficult for doctors to accurately diagnose the residual thyroid tissue in patients based on SPECT images. At present, the research on the classification of thyroid tissue residues after thyroidectomy is still in a blank state. This paper proposes a ResNet-18 fine-tuning method based on the convolutional neural network model. First, preprocess the SPECT images to improve the image quality and remove background interference. Secondly, use the preprocessed image samples to fine-tune the pretrained ResNet-18 model to obtain better features and finally use the Softmax classifier to diagnose the residual thyroid tissue. The method has been tested on SPECT images of 446 patients collected by local hospital and compared with the widely used lightweight network SqueezeNet model and ShuffleNetV2 model. Due to the small data set, this paper conducted 10 random grouping experiments. Each experiment divided the data set into training set and test set at a ratio of 3:1. The accuracy and sensitivity rates of the model proposed in this paper are 96.69% and 94.75%, which are significantly higher than other models (p < 0.05). The specificity and precision rates are 99.6% and 99.96%, respectively, and there is no significant difference compared with other models. (p > 0.05). The area under the curve of the proposed model, SqueezeNet, and ShuffleNetv2 are 0.988 (95% CI, 0.941-1.000), 0.898 (95% CI, 0.819-0.951) (p = 0.0257), and 0.885 (95% CI, 0.803-0.941) (p = 0.0057) (p < 0.05). We prove that this thyroid tissue residue classification system can be used as a computer-aided diagnosis method to effectively improve the diagnostic accuracy of thyroid tissue residues. While more accurately diagnosing patients with residual thyroid tissue in the body, we try our best to avoid the occurrence of overtreatment, which reflects its potential clinical application value.

Multi-Order Brain Functional Connectivity Network-Based Machine Learning Method for Recognition of Delayed Neurocognitive Recovery in Older Adults Undergoing Non-cardiac Surgery

Objectives: Delayed neurocognitive recovery (DNR) seriously affects the post-operative recovery of elderly surgical patients, but there is still a lack of effective methods to recognize high-risk patients with DNR. This study proposed a machine learning method based on a multi-order brain functional connectivity (FC) network to recognize DNR.

Method: Seventy-four patients who completed assessments were included in this study, in which 16/74 (21.6%) had DNR following surgery. Based on resting-state functional magnetic resonance imaging (rs-fMRI), we first constructed low-order FC networks of 90 brain regions by calculating the correlation of brain region signal changing in the time dimension. Then, we established high-order FC networks by calculating correlations among each pair of brain regions. Afterward, we built sparse representation-based machine learning model to recognize DNR on the extracted multi-order FC network features. Finally, an independent testing was conducted to validate the established recognition model.

Results: Three hundred ninety features of FC networks were finally extracted to identify DNR. After performing the independent-sample T test between these features and the categories, 15 features showed statistical differences (P < 0.05) and 3 features had significant statistical differences (P < 0.01). By comparing DNR and non-DNR patients’ brain region connection matrices, it is found that there are more connections among brain regions in DNR patients than in non-DNR patients. For the machine learning recognition model based on multi-feature combination, the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity of the classifier reached 95.61, 92.00, 66.67, and 100.00%, respectively.

Conclusion: This study not only reveals the significance of preoperative rs-fMRI in recognizing post-operative DNR in elderly patients but also establishes a promising machine learning method to recognize DNR.

Analysis of EPID Transmission Fluence Maps Using Machine Learning Models and CNN for Identifying Position Errors in the Treatment of GO Patients

Purpose

To find a suitable method for analyzing electronic portal imaging device (EPID) transmission fluence maps for the identification of position errors in the in vivo dose monitoring of patients with Graves' ophthalmopathy (GO).

Methods

Position errors combining 0-, 2-, and 4-mm errors in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions in the delivery of 40 GO patient radiotherapy plans to a human head phantom were simulated and EPID transmission fluence maps were acquired. Dose difference (DD) and structural similarity (SSIM) maps were calculated to quantify changes in the fluence maps. Three types of machine learning (ML) models that utilize radiomics features of the DD maps (ML 1 models), features of the SSIM maps (ML 2 models), and features of both DD and SSIM maps (ML 3 models) as inputs were used to perform three types of position error classification, namely a binary classification of the isocenter error (type 1), three binary classifications of LR, SI, and AP direction errors (type 2), and an eight-element classification of the combined LR, SI, and AP direction errors (type 3). Convolutional neural network (CNN) was also used to classify position errors using the DD and SSIM maps as input.

Results

The best-performing ML 1 model was XGBoost, which achieved accuracies of 0.889, 0.755, 0.778, 0.833, and 0.532 in the type 1, type 2-LR, type 2-AP, type 2-SI, and type 3 classification, respectively. The best ML 2 model was XGBoost, which achieved accuracies of 0.856, 0.731, 0.736, 0.949, and 0.491, respectively. The best ML 3 model was linear discriminant classifier (LDC), which achieved accuracies of 0.903, 0.792, 0.870, 0.931, and 0.671, respectively. The CNN achieved classification accuracies of 0.925, 0.833, 0.875, 0.949, and 0.689, respectively.

Conclusion

ML models and CNN using combined DD and SSIM maps can analyze EPID transmission fluence maps to identify position errors in the treatment of GO patients. Further studies with large sample sizes are needed to improve the accuracy of CNN.

[Establishment of a predictive model for central cervical lymph node metastasis of papillary thyroid carcinoma based on ACR TI-RADS score and evaluation of its diagnostic efficacy]

Objective:To establish a predictive model for central lymph node metastasis（CLNM） of papillary thyroid carcinoma（PTC） based on ACR TI-RADS grades（ATR model） and evaluate its diagnostic efficacy. Methods:A total of 319 patients with PTC diagnosed from January 2019 to May 2020 were included, including 366 nodules were used as the modeling cohort to construct the risk prediction model. A total of 105 PTC patients with 121 nodules from June to August 2020 were included as the external validation cohort. The C-index of the model was calculated and the Hosmer-Lemeshow goodness-of-fit test was performed to compare the diagnostic efficiency of ACR model and those conventional imaging models. Results:The ATR model, Y=-3.719+0.765×gender+1.094×multifocality+0.08×maximum diameter+0.266×ACR TI-RADS score. In the training set, validation set and external validation cohort, the model C-index was 0.758（95%CI: 0.699-0.817）, 0.717（95%CI: 0.619-0.815） and 0.756（95%CI: 0.671-0.840）, respectively. The Hosmer-Lemeshow goodness of fit test showed that the prediction rate of the model was consistent with the actual incidence rate（P=0.918; P=0.581; P=0.366）. With ≥0.434 as the diagnostic threshold, the model had the highest diagnostic efficacy （sensitivity: 86.0%, specificity: 56.3%, Youden index: 0.423）. In the external validation cohort, there was no significant difference between C-US and CT（P>0.05）. Compared with C-US and CT, the sensitivity（66.1% vs 16.1%, P<0.001; 66.1% vs 9.7%, P<0.001） and accuracy（68.6% vs 55.4%, P=0.041; 68.6% vs 52.9%, P=0.012） of ATR model were higher, and the negative predictive value was higher than that of CT（66.7% vs 50.9%, P=0.042）, but there was no difference between ATR model and C-US（66.7% vs 52.3%, P=0.066）; There was no significant difference among the three positive predictive values（70.7% vs 83.3%, P=0.211; 70.7% vs 85.7%, P=0.319）, but the specificity of the model was low （71.2% vs 96.6%, P=0.001; 71.2% vs 98.3%, P<0.001）. Conclusion:The predictive model based on ACR TI-RADS grades can predict CLNM of PTC more accurately and sensitively than traditional imaging examination.

A Comprehensive Review on Radiomics and Deep Learning for Nasopharyngeal Carcinoma Imaging

Nasopharyngeal carcinoma (NPC) is one of the most common malignant tumours of the head and neck, and improving the efficiency of its diagnosis and treatment strategies is an important goal. With the development of the combination of artificial intelligence (AI) technology and medical imaging in recent years, an increasing number of studies have been conducted on image analysis of NPC using AI tools, especially radiomics and artificial neural network methods. In this review, we present a comprehensive overview of NPC imaging research based on radiomics and deep learning. These studies depict a promising prospect for the diagnosis and treatment of NPC. The deficiencies of the current studies and the potential of radiomics and deep learning for NPC imaging are discussed. We conclude that future research should establish a large-scale labelled dataset of NPC images and that studies focused on screening for NPC using AI are necessary.

Lens Opacities Classification System III-based artificial intelligence program for automatic cataract grading

PURPOSE

To establish and validate an artificial intelligence (AI)-assisted automatic cataract grading program based on the Lens Opacities Classification System III (LOCSIII).

SETTING

Eye and Ear, Nose, and Throat (EENT) Hospital, Fudan University, Shanghai, China.

DESIGN

AI training.

METHODS

Advanced deep-learning algorithms, including Faster R-CNN and ResNet, were applied to the localization and analysis of the region of interest. An internal dataset from the EENT Hospital of Fudan University and an external dataset from the Pujiang Eye Study were used for AI training, validation, and testing. The datasets were automatically labeled on the AI platform in terms of the capture mode and cataract grading based on the LOCSIII system.

RESULTS

The AI program showed reliable capture mode recognition, grading, and referral capability for nuclear and cortical cataract grading. In the internal and external datasets, 99.4% and 100% of automatic nuclear grading, respectively, had an absolute prediction error of ≤ 1.0, with a satisfactory referral capability (area under the curve [AUC]: 0.983 for the internal dataset; 0.977 for the external dataset). 75.0% (internal dataset) and 93.5% (external dataset) of the automatic cortical grades had an absolute prediction error of ≤ 1.0, with AUCs of 0.855 and 0.795 for referral, respectively. Good consistency was observed between automatic and manual grading when both nuclear and cortical cataracts were evaluated. However, automatic grading of posterior subcapsular cataracts was impractical.

CONCLUSIONS

The AI program proposed in this study shows robust grading and diagnostic performance for both nuclear and cortical cataracts, based on LOCSIII.

Preoperative prediction of lymph node metastasis in patients with papillary thyroid carcinoma by an artificial intelligence algorithm

BACKGROUND

It is necessary to identify patients at risk of developing lymph node metastasis prior to papillary thyroid carcinoma (PTC) surgery. This can be challenging due to limiting factors, and an artificial intelligence algorithm may be a viable option.

OBJECTIVE

In this study, we aimed to evaluate whether combining an artificial intelligence algorithm (support vector machine and probabilistic neural network) and clinico-pathologic data can preoperatively predict lymph node metastasis of papillary thyroid carcinoma (PTC).

METHODS

We retrospectively examined 251 PTCs with lymph node metastasis and 194 PTCs without lymph node metastasis. The artificial intelligence algorithm included the support vector machine (SVM) and the probabilistic neural network (PNN).

RESULTS

The ACR TI-RADS (Thyroid Imaging, Reporting and Data System), number of tumours, no well-defined margin, lymph node status and rim calcification on ultrasonography (US), age, sex, tumour size, and presence of Hashimoto's thyroiditis were significantly more frequent among PTCs with central lymph node metastasis than those without metastasis (P<0.05). The PNN classifier revealed an F1 score of 0.88 on the central lymph node metastasis test set. The SVM classifier revealed an F1 score of 0.93 on the lateral lymph node metastasis test set. Our study demonstrates that combining artificial intelligence algorithms and clinico-pathologic data can effectively predict the lymph node metastasis of papillary thyroid carcinoma prior to surgery.

Magnetic resonance imaging (MRI) radiomics of papillary thyroid cancer (PTC): a comparison of predictive performance of multiple classifiers modeling to identify cervical lymph node metastases before surgery

PURPOSE

To compare predictive efficiency of multiple classifiers modeling and establish a combined magnetic resonance imaging (MRI) radiomics model for identifying lymph node (LN) metastases of papillary thyroid cancer (PTC) preoperatively.

MATERIALS AND METHODS

A retrospective analysis based on the preoperative MRI scans of 109 PTC patients including 77 patients with LN metastases and 32 patients without metastases was conducted, and we divided enroll cases into trained group and validation group. Radiomics signatures were selected from fat-suppressed T2-weighted MRI images, and the optimal characteristics were confirmed by spearman correlation test, hypothesis testing and random forest methods, and then, eight predictive models were constructed by eight classifiers. The receiver operating characteristic (ROC) curves analysis were performed to demonstrate the effectiveness of the models.

RESULTS

The area under the curve (AUC) of ROC based on MRI texture diagnosed LN status by naked eye was 0.739 (sensitivity = 0.571, specificity = 0.906). Based on the 5 optimal signatures, the best AUC of MRI radiomics model by logistics regression classifier had a considerable prediction performance with AUCs 0.805 in trained group and 0.760 in validation group, respectively, and a combination of best radiomics model with visual diagnosis of MRI texture had a high AUC as 0.969 (sensitivity = 0.938, specificity = 1.000), suggesting combined model had a preferable diagnostic efficiency in evaluating LN metastases of PTC.

CONCLUSION

Our combined radiomics model with visual diagnosis could be a potentially effective strategy to preoperatively predict LN metastases in PTC patients before clinical intervention.

Radiogenomic Analysis of Papillary Thyroid Carcinoma for Prediction of Cervical Lymph Node Metastasis: A Preliminary Study

Background

Papillary thyroid carcinoma (PTC) is characterized by frequent metastases to cervical lymph nodes (CLNs), and the presence of lymph node metastasis at diagnosis has a significant impact on the surgical approach. Therefore, we established a radiomic signature to predict the CLN status of PTC patients using preoperative thyroid ultrasound, and investigated the association between the radiomic features and underlying molecular characteristics of PTC tumors.

Methods

In total, 270 patients were enrolled in this prospective study, and radiomic features were extracted according to multiple guidelines. A radiomic signature was built with selected features in the training cohort and validated in the validation cohort. The total protein extracted from tumor samples was analyzed with LC/MS and iTRAQ technology. Gene modules acquired by clustering were chosen for their diagnostic significance. A radiogenomic map linking radiomic features to gene modules was constructed with the Spearman correlation matrix. Genes in modules related to metastasis were extracted for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, and a protein-protein interaction (PPI) network was built to identify the hub genes in the modules. Finally, the screened hub genes were validated by immunohistochemistry analysis.

Results

The radiomic signature showed good performance for predicting CLN status in training and validation cohorts, with area under curve of 0.873 and 0.831 respectively. A radiogenomic map was created with nine significant correlations between radiomic features and gene modules, and two of them had higher correlation coefficient. Among these, MEmeganta representing the upregulation of telomere maintenance via telomerase and cell-cell adhesion was correlated with ‘Rectlike’ and ‘deviation ratio of tumor tissue and normal thyroid gland’ which reflect the margin and the internal echogenicity of the tumor, respectively. MEblue capturing cell-cell adhesion and glycolysis was associated with feature ‘minimum calcification area’ which measures the punctate calcification. The hub genes of the two modules were identified by protein-protein interaction network. Immunohistochemistry validated that LAMC1 and THBS1 were differently expressed in metastatic and non-metastatic tissues (p=0.003; p=0.002). And LAMC1 was associated with feature ‘Rectlike’ and ‘deviation ratio of tumor and normal thyroid gland’ (p<0.001; p<0.001); THBS1 was correlated with ‘minimum calcification area’ (p<0.001).

Conclusions

The radiomic signature proposed here has the potential to noninvasively predict the CLN status in PTC patients. Merging imaging phenotypes with genomic data could allow noninvasive identification of the molecular properties of PTC tumors, which might support clinical decision making and personalized management.

A Comparative Analysis of Six Machine Learning Models Based on Ultrasound to Distinguish the Possibility of Central Cervical Lymph Node Metastasis in Patients With Papillary Thyroid Carcinoma

Current approaches to predict central cervical lymph node metastasis (CLNM) in patients with papillary thyroid carcinoma (PTC) have failed to identify patients who would benefit from preventive treatment. Machine learning has offered the opportunity to improve accuracy by comparing the different algorithms. We assessed which machine learning algorithm can best improve CLNM prediction. This retrospective study used routine ultrasound data of 1,364 PTC patients. Six machine learning algorithms were compared to predict the possibility of CLNM. Predictive accuracy was assessed by sensitivity, specificity, positive predictive value, negative predictive value, and the area under the curve (AUC). The patients were randomly split into the training (70%), validation (15%), and test (15%) data sets. Random forest (RF) led to the best diagnostic model in the test cohort (AUC 0.731 ± 0.036, 95% confidence interval: 0.664–0.791). The diagnostic performance of the RF algorithm was most dependent on the following five top-rank features: extrathyroidal extension (27.597), age (17.275), T stage (15.058), shape (13.474), and multifocality (12.929). In conclusion, this study demonstrated promise for integrating machine learning methods into clinical decision-making processes, though these would need to be tested prospectively.

US-Based Deep Learning Model for Differentiating Hepatocellular Carcinoma (HCC) From Other Malignancy in Cirrhotic Patients

The aim was to build a predictive model based on ultrasonography (US)-based deep learning model (US-DLM) and clinical features (Clin) for differentiating hepatocellular carcinoma (HCC) from other malignancy (OM) in cirrhotic patients. 112 patients with 120 HCCs and 60 patients with 61 OMs were included. They were randomly divided into training and test cohorts with a 4:1 ratio for developing and evaluating US-DLM model, respectively. Significant Clin predictors of OM in the training cohort were combined with US-DLM to build a nomogram predictive model (US-DLM+Clin). The diagnostic performance of US-DLM and US-DLM+Clin were compared with that of contrast enhanced magnetic resonance imaging (MRI) liver imaging and reporting system category M (MRI LR-M). US-DLM was the best independent predictor for evaluating OMs, followed by clinical information, including high cancer antigen 199 (CA199) level and female. The US-DLM achieved an AUC of 0.74 in the test cohort, which was comparable with that of MRI LR-M (AUC=0.84, p=0.232). The US-DLM+Clin for predicting OMs also had similar AUC value (0.81) compared with that of LR-M+Clin (0.83, p>0.05). US-DLM+Clin obtained a higher specificity, but a lower sensitivity, compared to that of LR-M +Clin (Specificity: 82.6% vs. 73.9%, p=0.007; Sensitivity: 78.6% vs. 92.9%, p=0.006) for evaluating OMs in the test set. The US-DLM+Clin model is valuable in differentiating HCC from OM in the setting of cirrhosis.

Predictive Value of Folate Receptor-Positive Circulating Tumor Cells for the Preoperative Diagnosis of Lymph Node Metastasis in Patients with Lung Adenocarcinoma

Noninvasive assessments of the risk of lymph node metastasis (LNM) in patients with lung adenocarcinoma (LAD) are of great value for selecting individualized treatment options. However, the diagnostic accuracies of different preoperative LN evaluation methods in routine clinical practice are not satisfactory. Here, an assessment to detect folate receptor (FR)-positive circulating tumor cells (CTCs) based on ligand-targeted enzyme-linked polymerization is established. FR-positive CTCs have the potential to improve the specificity and sensitivity of diagnosing LNM in lung cancer patients. The addition of CTC level improved the diagnostic efficiency of the initial prediction model that comprises other clinical parameters. A nomogram for predicting preoperative LNM is established, which showed good prediction and calibration capacities and achieved an average area under the curve of 0.786. Good correlations are observed between the CTC level and nodal classifications, such as the number of positive LNs and the ratio of the number of positive LNs to removed LNs (LN ratio or LNR). The ligand-targeted enzyme-linked polymerization-assisted assessment of CTCs enables noninvasive detection and has a useful predictive value for the preoperative diagnosis of LNM in patients with LAD.

Radiomics in Differentiated Thyroid Cancer and Nodules: Explorations, Application, and Limitations

Radiomics is an emerging technique that allows the quantitative extraction of high-throughput features from single or multiple medical images, which cannot be observed directly with the naked eye, and then applies to machine learning approaches to construct classification or prediction models. This method makes it possible to evaluate tumor status and to differentiate malignant from benign tumors or nodules in a more objective manner. To date, the classification and prediction value of radiomics in DTC patients have been inconsistent. Herein, we summarize the available literature on the classification and prediction performance of radiomics-based DTC in various imaging techniques. More specifically, we reviewed the recent literature to discuss the capacity of radiomics to predict lymph node (LN) metastasis, distant metastasis, tumor extrathyroidal extension, disease-free survival, and B-Raf proto-oncogene serine/threonine kinase (BRAF) mutation and differentiate malignant from benign nodules. This review discusses the application and limitations of the radiomics process, and explores its ability to improve clinical decision-making with the hope of emphasizing its utility for DTC patients.

Radiomic and radiogenomic modeling for radiotherapy: strategies, pitfalls, and challenges

The power of predictive modeling for radiotherapy outcomes has historically been limited by an inability to adequately capture patient-specific variabilities; however, next-generation platforms together with imaging technologies and powerful bioinformatic tools have facilitated strategies and provided optimism. Integrating clinical, biological, imaging, and treatment-specific data for more accurate prediction of tumor control probabilities or risk of radiation-induced side effects are high-dimensional problems whose solutions could have widespread benefits to a diverse patient population-we discuss technical approaches toward this objective. Increasing interest in the above is specifically reflected by the emergence of two nascent fields, which are distinct but complementary: radiogenomics, which broadly seeks to integrate biological risk factors together with treatment and diagnostic information to generate individualized patient risk profiles, and radiomics, which further leverages large-scale imaging correlates and extracted features for the same purpose. We review classical analytical and data-driven approaches for outcomes prediction that serve as antecedents to both radiomic and radiogenomic strategies. Discussion then focuses on uses of conventional and deep machine learning in radiomics. We further consider promising strategies for the harmonization of high-dimensional, heterogeneous multiomics datasets (panomics) and techniques for nonparametric validation of best-fit models. Strategies to overcome common pitfalls that are unique to data-intensive radiomics are also discussed.

Endoscopic Left Central Lymph Node Dissection of Thyroid Cancer: Safe, Feasible, and Relatively Easy

Purpose. The purpose of this study was to explore the feasibility of left central lymph node dissection (CLND) in endoscopic thyroidectomy via chest-breast approach (ETCB). Methods. Retrospective analysis of 57 cases of left CLND (group A) via ETCB, 35 cases of open left CLND (group B), and 90 cases of right CLND via ETCB (Group C) were performed from October 2014 to October 2019. Surgical data, complications, and follow-up data were compared among group A and group B, group A and group C, respectively. Results. There were no significant differences between group A and group B in intraoperative blood loss, tumor size, lymph node (LN) metastasis rate, dissected LN number, metastatic LN number, serum thyroglobulin (sTg), radioactive iodine uptake (RAIU), radioactive technetium uptake (RATU), radionuclide imaging of the residual area (RITRA), and radionuclide imaging of suspicious lymph node metastasis (RISLNM). There were no significant differences between group A and group C in age, operation time, intraoperative blood loss, postoperative hospital stay, tumor size, LN metastasis rate, dissected LN number, metastatic LN number, hypoparathyroidism, sTg, RAIU, RATU, RITRA, and RISLNM. There were 5 cases of temporary recurrent laryngeal nerve (RLN) palsy and 1 case of recurrence in group C. Besides, 1 case of lymphatic leakage was in group A. Conclusion. For selected cases, endoscopic left CLND is safe, feasible, efficient, and more easier than endoscopic right CLND.

Machine learning to identify lymph node metastasis from thyroid cancer in patients undergoing contrast-enhanced CT studies

INTRODUCTION

We compared the diagnostic performance of morphological methods such as the major axis, the minor axis, the volume and sphericity and of machine learning with texture analysis in the identification of lymph node metastasis in patients with thyroid cancer who had undergone contrast-enhanced CT studies.

METHODS

We sampled 772 lymph nodes with histology defined tissue types (84 metastatic and 688 benign lymph nodes) that were visualised on CT images of 117 patients. A support vector machine (SVM), free programming software (Python), and the scikit-learn machine learning library were used to discriminate metastatic-from benign lymph nodes. We assessed 96 texture and 4 morphological features (major axis, minor axis, volume, sphericity) that were reported useful for the differentiation between metastatic and benign lymph nodes on CT images. The area under the curve (AUC) obtained by receiver operating characteristic analysis of univariate logistic regression and SVM classifiers were calculated for the training and testing datasets.

RESULTS

The AUC for all classifiers in training and testing datasets was 0.96 and 0.86, at the SVM for machine learning. When we applied conventional methods to the training and testing datasets, the AUCs were 0.63 and 0.48 for the major axis, 0.70 and 0.44 for the minor axis, 0.66 and 0.43 for the volume, and 0.69 and 0.54 for sphericity, respectively. The SVM using texture features yielded significantly higher AUCs than univariate logistic regression models using morphological features (p = 0.001).

CONCLUSION

For the identification of metastatic lymph nodes from thyroid cancer on contrast-enhanced CT images, machine learning combined with texture analysis was superior to conventional diagnostic methods with the morphological parameters.

IMPLICATIONS FOR PRACTICE

Our findings suggest that in patients with thyroid cancer and suspected lymph node metastasis who undergo contrast-enhanced CT studies, machine learning using texture analysis is high diagnostic value for the identification of metastatic lymph nodes.

Artificial Intelligence for Personalized Medicine in Thyroid Cancer: Current Status and Future Perspectives

Thyroid cancers (TC) have increasingly been detected following advances in diagnostic methods. Risk stratification guided by refined information becomes a crucial step toward the goal of personalized medicine. The diagnosis of TC mainly relies on imaging analysis, but visual examination may not reveal much information and not enable comprehensive analysis. Artificial intelligence (AI) is a technology used to extract and quantify key image information by simulating complex human functions. This latent, precise information contributes to stratify TC on the distinct risk and drives tailored management to transit from the surface (population-based) to a point (individual-based). In this review, we started with several challenges regarding personalized care in TC, for example, inconsistent rating ability of ultrasound physicians, uncertainty in cytopathological diagnosis, difficulty in discriminating follicular neoplasms, and inaccurate prognostication. We then analyzed and summarized the advances of AI to extract and analyze morphological, textural, and molecular features to reveal the ground truth of TC. Consequently, their combination with AI technology will make individual medical strategies possible.

Best Achievements in Clinical Thyroidology in 2020

This review highlights the most interesting research in thyroidology conducted in 2020. The publications of interest discussed below dealt with the following topics: thyroid dysfunction, risk of thyroid cancer, molecular diagnostics and new therapeutics for thyroid cancer, and thyroid disease in the coronavirus disease 2019 pandemic era.

Features of Lymph Node Metastasis and Structural Recurrence in Papillary Thyroid Carcinoma Located in the Upper Portion of the Thyroid: A Retrospective Cohort Study

BACKGROUND

This study aims to reveal the features of lymph node metastasis (LNM) and recurrence in papillary thyroid carcinoma (PTC) tumors located in the upper portion of the thyroid.

METHODS

A total of 1075 PTC patients were retrospectively reviewed, including 314 patients with a tumor in the upper portion of the thyroid. Another 103 PTC patients with upper portion diagonsis from three clinical centers were included for external validation.

RESULTS

The results showed no difference between the patients with a tumor in the upper portion of the thyroid and those with a tumor in the non-upper portion in terms of overall LNM rates. However, patients with a tumor in the upper portion were significantly more prone to LLNM and exhibited a significantly worse recurrence outcome than those with a tumor in other subregions. Multivariate analysis showed that four factors-age no more than 40, maximum tumor diameter no less than1.0 cm, the presence of thyroid capsular invasion, and tumor with ipsilateral nodular goiter-were independent risk factors for LLNM of the tumor in the upper thyroid. A predictive risk-scoring model was established based on these factors.

CONCLUSIONS

Patients with PTC located in the upper portion may have an exclusive lymphatic drainage pathway to the lateral neck region and are more prone to suffer from LLNM and tumor recurrence than those with a tumor located in other subregions. A new postoperative strategy selection flow chart was established based on our newly created risk-scoring model that can effectively predict the individualized possibility of LLNM for PTC patients with a tumor in the upper portion.

Computed Tomography-Based Radiomics Model to Predict Central Cervical Lymph Node Metastases in Papillary Thyroid Carcinoma: A Multicenter Study

OBJECTIVES

This study aimed to develop a computed tomography (CT)-based radiomics model to predict central lymph node metastases (CLNM) preoperatively in patients with papillary thyroid carcinoma (PTC).

METHODS

In this retrospective study, 678 patients with PTC were enrolled from Yantai Yuhuangding Hot3spital (n=605) and the Affiliated Hospital of Binzhou Medical University (n=73) within August 2010 to December 2020. The patients were randomly divided into a training set (n=423), an internal test set (n=182), and an external test set (n=73). Radiomics features of each patient were extracted from preoperative plain scan and contrast-enhanced CT images (arterial and venous phases). One-way analysis of variance (ANOVA) and least absolute shrinkage and selection operator algorithm were used for feature selection. The K-nearest neighbor, logistics regression, decision tree, linear-support vector machine (linear-SVM), Gaussian-SVM, and polynomial-SVM algorithms were used to establish radiomics models for CLNM prediction. The clinical risk factors were selected by ANOVA and multivariate logistic regression. Incorporated with clinical risk factors, a combined radiomics model was established for the preoperative prediction of CLNM in patients with PTCs. The performance of the combined radiomics model was evaluated using the receiver operating characteristic (ROC) and calibration curves in the training and test sets. The clinical usefulness was evaluated through decision curve analysis (DCA).

RESULTS

A total of 4227 radiomic features were extracted from the CT images of each patient, and 14 non-zero coefficient features associated with CLNM were selected. Four clinical variables (sex, age, tumor diameter, and CT-reported lymph node status) were significantly associated with CLNM. Linear-SVM led to the best prediction model, which incorporated radiomic features and clinical risk factors. Areas under the ROC curves of 0.747 (95% confidence interval [CI] 0.706-0.782), 0.710 (95% CI 0.634-0.786), and 0.764 (95% CI 0.654-0.875) were obtained in the training, internal, and external test sets, respectively. The linear-SVM algorithm also showed better sensitivity (0.702 [95% CI 0.600-0.790] vs. 0.477 [95% CI 0.409-0.545]) and accuracy (0.670 [95% CI 0.600-0.738] vs. 0.642 [95% CI 0.569-0.712]) than an experienced radiologist in the internal test set in the combined radiomics model. The calibration plot reflected a favorable agreement between the actual and estimated probabilities of CLNM. The DCA indicated the clinical usefulness of the combined radiomics model.

CONCLUSION

The combined radiomics model is a non-invasive preoperative tool that incorporates radiomic features and clinical risk factors to predict CLNM in patients with PTC.