Performance of large language models on benign prostatic hyperplasia frequently asked questions.

BACKGROUND: Benign prostatic hyperplasia (BPH) is a common condition, yet it is challenging for the average BPH patient to find credible and accurate information about BPH. Our goal is to evaluate and compare the accuracy and reproducibility of large language models (LLMs), including ChatGPT-3.5, ChatGPT-4, and the New Bing Chat in responding to a BPH frequently asked questions (FAQs) questionnaire. METHODS: A total of 45 questions related to BPH were categorized into basic and professional knowledge. Three LLM-ChatGPT-3.5, ChatGPT-4, and New Bing Chat-were utilized to generate responses to these questions. Responses were graded as comprehensive, correct but inadequate, mixed with incorrect/outdated data, or completely incorrect. Reproducibility was assessed by generating two responses for each question. All responses were reviewed and judged by experienced urologists. RESULTS: All three LLMs exhibited high accuracy in generating responses to questions, with accuracy rates ranging from 86.7% to 100%. However, there was no statistically significant difference in response accuracy among the three (p > 0.017 for all comparisons). Additionally, the accuracy of the LLMs' responses to the basic knowledge questions was roughly equivalent to that of the specialized knowledge questions, showing a difference of less than 3.5% (GPT-3.5: 90% vs. 86.7%; GPT-4: 96.7% vs. 95.6%; New Bing: 96.7% vs. 93.3%). Furthermore, all three LLMs demonstrated high reproducibility, with rates ranging from 93.3% to 97.8%. CONCLUSIONS: ChatGPT-3.5, ChatGPT-4, and New Bing Chat offer accurate and reproducible responses to BPH-related questions, establishing them as valuable resources for enhancing health literacy and supporting BPH patients in conjunction with healthcare professionals.

Generating lymphoma ultrasound image description with transformer model.

Lymphoma, the most prevalent hematologic tumor originating from the lymphatic hematopoietic system, can be accurately diagnosed using high-resolution ultrasound. Microscopic ultrasound performance enables clinicians to identify suspected tumors and subsequently obtain a definitive pathological diagnosis through puncture biopsy. However, the complex and diverse ultrasonographic manifestations of lymphoma pose challenges for accurate characterization by sonographers. To address these issues, this study proposes a Transformer-based model for generating descriptive ultrasound images of lymphoma, aiming to provide auxiliary guidance for ultrasound doctors during screening procedures. Specifically, deep stable learning is integrated into the model to eliminate feature dependencies by training sample weights. Additionally, a memory module is incorporated into the model decoder to enhance semantic information modeling in descriptions and utilize learned semantic tree branch structures for more detailed image depiction. Experimental results on an ultrasonic diagnosis dataset from Shanghai Ruijin Hospital demonstrate that our proposed model outperforms relevant methods in terms of prediction performance.

Ultrasound Viscosity Imaging in Breast Lesions: A Multicenter Prospective Study.

RATIONALE AND OBJECTIVES: To explore and validate the clinical value of ultrasound (US) viscosity imaging in differentiating breast lesions by combining with BI-RADS, and then comparing the diagnostic performances with BI-RADS alone. MATERIALS AND METHODS: This multicenter, prospective study enrolled participants with breast lesions from June 2021 to November 2022. A development cohort (DC) and validation cohort (VC) were established. Using histological results as reference standard, the viscosity-related parameter with the highest area under the receiver operating curve (AUC) was selected as the optimal one. Then the original BI-RADS would upgrade or not based on the value of this parameter. Finally, the results were validated in the VC and total cohorts. In the DC, VC and total cohorts, all breast lesions were divided into the large lesion, small lesion and overall groups respectively. RESULTS: A total of 639 participants (mean age, 46 years ± 14) with 639 breast lesions (372 benign and 267 malignant lesions) were finally enrolled in this study including 392 participants in the DC and 247 in the VC. In the DC, the optimal viscosity-related parameter in differentiating breast lesions was calculated to be A'-S2-Vmax, with the AUC of 0.88 (95% CI: 0.84, 0.91). Using > 9.97 Pa.s as the cutoff value, the BI-RADS was then modified. The AUC of modified BI-RADS significantly increased from 0.85 (95% CI: 0.81, 0.88) to 0.91 (95% CI: 0.87, 0.93), 0.85 (95% CI: 0.80, 0.89) to 0.90 (95% CI: 0.85, 0.93) and 0.85 (95% CI: 0.82, 0.87) to 0.90 (95% CI: 0.88, 0.92) in the DC, VC and total cohorts respectively (P < .05 for all). CONCLUSION: The quantitative viscous parameters evaluated by US viscosity imaging contribute to breast cancer diagnosis when combined with BI-RADS.

Transforming free-text radiology reports into structured reports using ChatGPT: A study on thyroid ultrasonography.

PURPOSE: The importance of structured radiology reports has been fully recognized, as they facilitate efficient data extraction and promote collaboration among healthcare professionals. Our purpose is to assess the accuracy and reproducibility of ChatGPT, a large language model, in generating structured thyroid ultrasound reports. METHODS: This is a retrospective study that includes 184 nodules in 136 thyroid ultrasound reports from 136 patients. ChatGPT-3.5 and ChatGPT-4.0 were used to structure the reports based on ACR-TIRADS guidelines. Two radiologists evaluated the responses for quality, nodule categorization accuracy, and management recommendations. Each text was submitted twice to assess the consistency of the nodule classification and management recommendations. RESULTS: On 136 ultrasound reports from 136 patients (mean age, 52 years ± 12 [SD]; 61 male), ChatGPT-3.5 generated 202 satisfactory structured reports, while ChatGPT-4.0 only produced 69 satisfactory structured reports (74.3 % vs. 25.4 %, odds ratio (OR) = 8.490, 95 %CI: 5.775-12.481, p < 0.001). ChatGPT-4.0 outperformed ChatGPT-3.5 in categorizing thyroid nodules, with an accuracy of 69.3 % compared to 34.5 % (OR = 4.282, 95 %CI: 3.145-5.831, p < 0.001). ChatGPT-4.0 also provided more comprehensive or correct management recommendations than ChatGPT-3.5 (OR = 1.791, 95 %CI: 1.297-2.473, p < 0.001). Finally, ChatGPT-4.0 exhibits higher consistency in categorizing nodules compared to ChatGPT-3.5 (ICC = 0.732 vs. ICC = 0.429), and both exhibited moderate consistency in management recommendations (ICC = 0.549 vs ICC = 0.575). CONCLUSIONS: Our study demonstrates the potential of ChatGPT in transforming free-text thyroid ultrasound reports into structured formats. ChatGPT-3.5 excels in generating structured reports, while ChatGPT-4.0 shows superior accuracy in nodule categorization and management recommendations.

Immunogenicity and effectiveness of an mRNA therapeutic vaccine for HPV-related malignancies.

Human papillomavirus (HPV) infections account for several human cancers. There is an urgent need to develop therapeutic vaccines for targeting preexisting high-risk HPV (such as HPV 16 and 18) infections and lesions, which are insensitive to preventative vaccines. In this study, we developed a lipid nanoparticle-formulated mRNA-based HPV therapeutic vaccine (mHTV), mHTV-02, targeting the E6/E7 of HPV16 and HPV-18. mHTV-02 dramatically induced antigen-specific cellular immune response and robust memory T-cell immunity in mice, besides significant CD8+ T-cell infiltration and cytotoxicity in TC-1 tumors expressing HPV E6/E7, resulting in tumor regression and prolonged survival in mice. Moreover, evaluation of routes of administration found that intramuscular or intratumoral injection of mHTV-02 displayed significant therapeutic effects. In contrast, intravenous delivery of the vaccine barely showed any benefit in reducing tumor size or improving animal survival. These data together support mHTV-02 as a candidate therapeutic mRNA vaccine via specific administration routes for treating malignancies caused by HPV16 or HPV18 infections.

An efficient framework for lesion segmentation in ultrasound images using global adversarial learning and region-invariant loss.

Lesion segmentation in ultrasound images is an essential yet challenging step for early evaluation and diagnosis of cancers. In recent years, many automatic CNN-based methods have been proposed to assist this task. However, most modern approaches often lack capturing long-range dependencies and prior information making it difficult to identify the lesions with unfixed shapes, sizes, locations, and textures. To address this, we present a novel lesion segmentation framework that guides the model to learn the global information about lesion characteristics and invariant features (e.g., morphological features) of lesions to improve the segmentation in ultrasound images. Specifically, the segmentation model is guided to learn the characteristics of lesions from the global maps using an adversarial learning scheme with a self-attention-based discriminator. We argue that under such a lesion characteristics-based guidance mechanism, the segmentation model gets more clues about the boundaries, shapes, sizes, and positions of lesions and can produce reliable predictions. In addition, as ultrasound lesions have different textures, we embed this prior knowledge into a novel region-invariant loss to constrain the model to focus on invariant features for robust segmentation. We demonstrate our method on one in-house breast ultrasound (BUS) dataset and two public datasets (i.e., breast lesion (BUS B) and thyroid nodule from TNSCUI2020). Experimental results show that our method is specifically suitable for lesion segmentation in ultrasound images and can outperform the state-of-the-art approaches with Dice of 0.931, 0.906, and 0.876, respectively. The proposed method demonstrates that it can provide more important information about the characteristics of lesions for lesion segmentation in ultrasound images, especially for lesions with irregular shapes and small sizes. It can assist the current lesion segmentation models to better suit clinical needs.

Secretory CAZymes profile and GH19 enzymes analysis of Corallococcus silvisoli c25j21.

Extracellular enzymes play important roles in myxobacteria degrading macromolecules and preying on other microorganisms. Glycoside hydrolases 19 (GH19) are widely present in myxobacteria, but their evolution and biological functions have not been fully elucidated. Here we investigated the comparative secretory proteome of Corallococcus silvisoli c25j21 in the presence of cellulose and chitin. A total of 313 proteins were detected, including 16 carbohydrate-active enzymes (CAZymes), 7 of which were induced by cellulose or chitin, such as GH6, GH13, GH19, AA4, and CBM56. We further analyzed the sequence and structural characteristics of its three GH19 enzymes to understand their potential functions. The results revealed that myxobacterial GH19 enzymes are evolutionarily divided into two clades with different appended modules, and their different amino acid compositions in the substrate binding pockets lead to the differences in molecular surface electrostatic potentials, which may, in turn, affect their substrate selectivity and biological functions. Our study is helpful for further understanding the biological functions and catalytic mechanisms of myxobacterial CAZymes.

Comparison of the efficacy and safety of ultrasound-guided radiofrequency ablation and microwave ablation for the treatment of unifocal papillary thyroid microcarcinoma: a retrospective study.

PURPOSE: This study aimed to compare the efficacy and safety of ultrasound-guided RFA and MWA in the treatment of unifocal PTMC. METHODS: This retrospective study included 512 patients with 512 unifocal papillary thyroid microcarcinomas (PTMCs) who underwent RFA (n = 346) and MWA (n = 166) between January 2021 and December 2021. The volumes of the ablation areas were measured during follow-up, and the volume reduction rates were evaluated. The ablation duration, volume of hydrodissection, and ablation-related complications were also compared between the groups. RESULTS: All lesions received complete ablation and no local or distant recurrences were observed in the two groups. A larger volume of isolation liquid was used for RFA than for MWA (p = 0.000). Hoarseness occurred in seven patients who underwent RFA (p = 0.102). At the 1-week follow-up, the mean volume of the areas ablated by RFA was smaller than that of the areas ablated by MWA (p = 0.049). During follow-ups at months 3, 9, 12, 15, and 18, the mean volumes of the ablated areas were larger in the RFA group than in the MWA group (all, p < 0.05). The mean volume of the ablated lesions increased slightly at the 1-week follow-up and then decreased at 1 month after ablation in both groups. The absorption curve of the ablated lesions in the RFA group was similar to that in the MWA group. CONCLUSIONS: RFA and MWA are both efficient and safe methods for treating unifocal PTMC. They may be alternative techniques for patients who are not eligible or are unwilling to undergo surgery.

Volume reduction rate of radiofrequency ablation in ≤ 2 cm Bethesda IV thyroid nodules.

OBJECTIVE: This prospective observational study aimed to evaluate the efficacy of radiofrequency ablation (RFA) in treating ≤ 2 cm thyroid nodules with Bethesda IV cytology and C-TIRADS 4A categorization. Additionally, the factors influencing the completed absorption of ablation (CAA) were examined. METHODS: A total of 62 cases with 62 nodules underwent ultrasound-guided RFA and were included in the study. The volume reduction rate (VRR), CAA, and incomplete absorption of ablation (IAA) were assessed at the 1st, 3rd, 6th, and subsequent 6-month follow-ups. Clinical and ultrasound features were compared between the CAA and IAA groups at the 12th month follow-up. RESULTS: The average VRR at the 1st, 3rd, 6th, 12th month, and last follow-up were -88.6%, 16.0%, 59.7%, 82.0%, and 98.2%, respectively. More than half of the nodules achieved a 90% VRR after 1 year of RFA, with 88.7% demonstrating CAA at the end of the study (follow-up duration of 14 to 63 months). Nodules with grade 3 vascularity and those associated with chronic thyroiditis showed delayed CAA at the 12th month follow-up (p = 0.036 and 0.003, respectively). CONCLUSION: RFA is an effective technique for treating ≤ 2 cm thyroid nodules with Bethesda IV cytology and C-TIRADS 4A categorization. Nodules with grade 3 blood supply and patients with chronic thyroiditis exhibited an impact on the completed absorption following RFA. CLINICAL RELEVANCE STATEMENT: Our study has shown that radiofrequency ablation is an effective treatment for ≤ 2 cm thyroid nodules classified as Bethesda IV cytology. However, we identified that high vascularity of the nodule and chronic thyroiditis are adverse factors affecting the completed absorption of the ablation. KEY POINTS: •Radiofrequency ablation (RFA) is an effective technique for treatment of ≤ 2 cm Bethesda IV category thyroid nodules. •Higher blood supply and chronic thyroiditis influence the completed absorption after RFA.

Generalizability and Diagnostic Performance of AI Models for Thyroid US.

Background Artificial intelligence (AI) models have improved US assessment of thyroid nodules; however, the lack of generalizability limits the application of these models. Purpose To develop AI models for segmentation and classification of thyroid nodules in US using diverse data sets from nationwide hospitals and multiple vendors, and to measure the impact of the AI models on diagnostic performance. Materials and Methods This retrospective study included consecutive patients with pathologically confirmed thyroid nodules who underwent US using equipment from 12 vendors at 208 hospitals across China from November 2017 to January 2019. The detection, segmentation, and classification models were developed based on the subset or complete set of images. Model performance was evaluated by precision and recall, Dice coefficient, and area under the receiver operating characteristic curve (AUC) analyses. Three scenarios (diagnosis without AI assistance, with freestyle AI assistance, and with rule-based AI assistance) were compared with three senior and three junior radiologists to optimize incorporation of AI into clinical practice. Results A total of 10 023 patients (median age, 46 years [IQR 37-55 years]; 7669 female) were included. The detection, segmentation, and classification models had an average precision, Dice coefficient, and AUC of 0.98 (95% CI: 0.96, 0.99), 0.86 (95% CI: 0.86, 0.87), and 0.90 (95% CI: 0.88, 0.92), respectively. The segmentation model trained on the nationwide data and classification model trained on the mixed vendor data exhibited the best performance, with a Dice coefficient of 0.91 (95% CI: 0.90, 0.91) and AUC of 0.98 (95% CI: 0.97, 1.00), respectively. The AI model outperformed all senior and junior radiologists (P < .05 for all comparisons), and the diagnostic accuracies of all radiologists were improved (P < .05 for all comparisons) with rule-based AI assistance. Conclusion Thyroid US AI models developed from diverse data sets had high diagnostic performance among the Chinese population. Rule-based AI assistance improved the performance of radiologists in thyroid cancer diagnosis. © RSNA, 2023 Supplemental material is available for this article.

Markedly hypoechoic: a new definition improves the diagnostic performance of thyroid ultrasound.

OBJECTIVES: To determine the contribution of a modified definition of markedly hypoechoic in the differential diagnosis of thyroid nodules. METHODS: A total of 1031 thyroid nodules were included in this retrospective multicenter study. All of the nodules were examined with US before surgery. The US features of the nodules were evaluated, in particular, the classical markedly hypoechoic and modified markedly hypoechoic (decreased or similar echogenicity relative to the adjacent strap muscles). The sensitivity, specificity, and AUC of classical/modified markedly hypoechoic and the corresponding ACR-TIRADS, EU-TIRADS, and C-TIRADS categories were calculated and compared. The inter- and intraobserver variability in the evaluation of the main US features of the nodules was assessed. RESULTS: There were 264 malignant nodules and 767 benign nodules. Compared with classical markedly hypoechoic as a diagnostic criterion for malignancy, using modified markedly hypoechoic as the criterion resulted in a significant increase in sensitivity (28.03% vs. 63.26%) and AUC (0.598 vs. 0.741), despite a significant decrease in specificity (91.53% vs. 84.88%) (p < 0.001 for all). Compared to the AUC of the C-TIRADS with the classical markedly hypoechoic, the AUC of the C-TIRADS with the modified markedly hypoechoic increased from 0.878 to 0.888 (p = 0.01); however, the AUCs of the ACR-TIRADS and EU-TIRADS did not change significantly (p > 0.05 for both). There was substantial interobserver agreement (κ = 0.624) and perfect intraobserver agreement (κ = 0.828) for the modified markedly hypoechoic. CONCLUSION: The modified definition of markedly hypoechoic resulted in a significantly improved diagnostic efficacy in determining malignant thyroid nodules and may improve the diagnostic performance of the C-TIRADS. CLINICAL RELEVANCE STATEMENT: Our study found that, compared with the original definition, modified markedly hypoechoic significantly improved the diagnostic performance in differentiating malignant from benign thyroid nodules and the predictive efficacy of the risk stratification systems. KEY POINTS: • Compared with the classical markedly hypoechoic as a diagnostic criterion for malignancy, the modified markedly hypoechoic resulted in a significant increase in sensitivity and AUC. • The C-TIRADS with the modified markedly hypoechoic achieved higher AUC and specificity than that with the classical markedly hypoechoic (p = 0.01 and < 0.001, respectively).

Bacillus velezensis RC116 Inhibits the Pathogens of Bacterial Wilt and Fusarium Wilt in Tomato with Multiple Biocontrol Traits.

Soil-borne plant diseases seriously threaten the tomato industry worldwide. Currently, eco-friendly biocontrol strategies have been increasingly considered as effective approaches to control the incidence of disease. In this study, we identified bacteria that could be used as biocontrol agents to mitigate the growth and spread of the pathogens causing economically significant diseases of tomato plants, such as tomato bacterial wilt and tomato Fusarium wilt. Specifically, we isolated a strain of Bacillus velezensis (RC116) from tomato rhizosphere soil in Guangdong Province, China, with high biocontrol potential and confirmed its identity using both morphological and molecular approaches. RC116 not only produced protease, amylase, lipase, and siderophores but also secreted indoleacetic acid, and dissolved organophosphorus in vivo. Moreover, 12 Bacillus biocontrol maker genes associated with antibiotics biosynthesis could be amplified in the RC116 genome. Extracellular secreted proteins of RC116 also exhibited strong lytic activity against Ralstonia solanacearum and Fusarium oxysporum f. sp. Lycopersici. Pot experiments showed that the biocontrol efficacy of RC116 against tomato bacteria wilt was 81%, and consequently, RC116 significantly promoted the growth of tomato plantlets. Based on these multiple biocontrol traits, RC116 is expected to be developed into a broad-spectrum biocontrol agent. Although several previous studies have examined the utility of B. velezensis for the control of fungal diseases, few studies to date have evaluated the utility of B. velezensis for the control of bacterial diseases. Our study fills this research gap. Collectively, our findings provide new insights that will aid the control of soil-borne diseases, as well as future studies of B. velezensis strains.

Virtual elastography ultrasound via generative adversarial network for breast cancer diagnosis.

Elastography ultrasound (EUS) imaging is a vital ultrasound imaging modality. The current use of EUS faces many challenges, such as vulnerability to subjective manipulation, echo signal attenuation, and unknown risks of elastic pressure in certain delicate tissues. The hardware requirement of EUS also hinders the trend of miniaturization of ultrasound equipment. Here we show a cost-efficient solution by designing a deep neural network to synthesize virtual EUS (V-EUS) from conventional B-mode images. A total of 4580 breast tumor cases were collected from 15 medical centers, including a main cohort with 2501 cases for model establishment, an external dataset with 1730 cases and a portable dataset with 349 cases for testing. In the task of differentiating benign and malignant breast tumors, there is no significant difference between V-EUS and real EUS on high-end ultrasound, while the diagnostic performance of pocket-sized ultrasound can be improved by about 5% after V-EUS is equipped.

Whole-genome sequencing of a biocontrol Myxococcus xanthus R31 isolate and comparative genomic analysis.

Tomato bacterial wilt (TBW) caused by Ralstonia solanacearum is one of the most destructive soil-borne diseases. Myxococcus xanthus R31, isolated from healthy tomato rhizosphere soil using the R. solanacearum baiting method, exhibiting good biocontrol efficacy against TBW. However, the genomic information and evolutionary features of R31 are largely unclear. Here, the high-quality genome assembly of R31 was presented. Using Nanopore sequencing technology, we assembled the 9.25 Mb complete genome of R31 and identified several extracellular enzyme proteins, including carbohydrate-active enzymes (CAZymes) and peptidases. We also performed a comparative genome analysis of R31 and 17 other strains of M. xanthus with genome sequences in the NCBI database to gain insights into myxobacteria predation and genome size expansion. Average nucleotide identity and digital DNA-DNA hybridization calculation and phylogenetic analysis indicated that R31 was closely related to the species M. xanthus. Further comparative genomics analysis suggested that, in addition to characteristics of predatory microorganisms, R31 contains many strain-specific genes, which may provide a genetic basis for its proficient predatory ability. This study provides new insights into R31 and other closely related species and facilitates studies using genetic approaches to further elucidate the predation mechanism of myxobacteria.

Hrip1 enhances tomato resistance to yellow leaf curl virus by manipulating the phenylpropanoid biosynthesis and plant hormone pathway.

Tomato yellow leaf curl virus (TYLCV) causes tremendous losses of tomato worldwide. An elicitor Hrip1, which produced by Alternaria tenuissima, can serve as a pathogen-associated molecular patterns (PAMPs) to trigger the immune defense response in Nicotiana benthamiana. Here, we show that Hrip1 can be targeted to the extracellular space and significantly delayed the development of symptoms caused by TYLCV in tomato. In basis of RNA-seq profiling, we find that 1621 differential expression genes (DEGs) with the opposite expression patterns are enriched in plant response to biotic stress between Hrip1 treatment and TYLCV infection of tomato. Thirty-two known differential expression miRNAs with the opposite expression patterns are identified by small RNA sequencing and the target genes of these miRNAs are significantly enriched in phenylpropanoid biosynthesis, plant hormone signal transduction and peroxisome. Based on the Pearson correlation analysis, 13 negative and 21 positive correlations are observed between differential expression miRNAs and DEGs. These miRNAs, which act as a key mediator of tomato resistance to TYLCV induced by Hrip1, regulate the expression of phenylpropanoid biosynthesis and plant hormone signal transduction-related genes. Taken together, our results provide an insight into tomato resistance to TYLCV induced by PAMP at transcriptional and posttranscriptional levels. Supplementary Information: The online version contains supplementary material available at 10.1007/s13205-022-03426-6.

A Disturbed Siderophore Transport Inhibits Myxobacterial Predation.

BACKGROUND: Understanding the intrinsic mechanisms of bacterial competition is a fundamental question. Iron is an essential trace nutrient that bacteria compete for. The most prevalent manner for iron scavenging is through the secretion of siderophores. Although tremendous efforts have focused on elucidating the molecular mechanisms of siderophores biosynthesis, export, uptake, and regulation of siderophores, the ecological aspects of siderophore-mediated competition are not well understood. METHODS: We performed predation and bacterial competition assays to investigate the function of siderophore transport on myxobacterial predation. RESULTS: Deletion of msuB, which encodes an iron chelate uptake ABC transporter family permease subunit, led to a reduction in myxobacterial predation and intracellular iron, but iron deficiency was not the predominant reason for the decrease in the predation ability of the ∆msuB mutant. We further confirmed that obstruction of siderophore transport decreased myxobacterial predation by investigating the function of a non-ribosomal peptide synthetase for siderophore biosynthesis, a TonB-dependent receptor, and a siderophore binding protein in M. xanthus. Our results showed that the obstruction of siderophores transport decreased myxobacterial predation ability through the downregulation of lytic enzyme genes, especially outer membrane vesicle (OMV)-specific proteins. CONCLUSIONS: This work provides insight into the mechanism of siderophore-mediated competition in myxobacteria.

Novel Human Artificial Intelligence Hybrid Framework Pinpoints Thyroid Nodule Malignancy and Identifies Overlooked Second-Order Ultrasonographic Features.

We present a Human Artificial Intelligence Hybrid (HAIbrid) integrating framework that reweights Thyroid Imaging Reporting and Data System (TIRADS) features and the malignancy score predicted by a convolutional neural network (CNN) for nodule malignancy stratification and diagnosis. We defined extra ultrasonographical features from color Doppler images to explore malignancy-relevant features. We proposed Gated Attentional Factorization Machine (GAFM) to identify second-order interacting features trained via a 10 fold distribution-balanced stratified cross-validation scheme on ultrasound images of 3002 nodules all finally characterized by postoperative pathology (1270 malignant ones), retrospectively collected from 131 hospitals. Our GAFM-HAIbrid model demonstrated significant improvements in Area Under the Curve (AUC) value (p-value < 10−5), reaching about 0.92 over the standalone CNN (~0.87) and senior radiologists (~0.86), and identified a second-order vascularity localization and morphological pattern which was overlooked if only first-order features were considered. We validated the advantages of the integration framework on an already-trained commercial CNN system and our findings using an extra set of ultrasound images of 500 nodules. Our HAIbrid framework allows natural integration to clinical workflow for thyroid nodule malignancy risk stratification and diagnosis, and the proposed GAFM-HAIbrid model may help identify novel diagnosis-relevant second-order features beyond ultrasonography.

Multi-Attribute Attention Network for Interpretable Diagnosis of Thyroid Nodules in Ultrasound Images.

Ultrasound (US) is the primary imaging technique for the diagnosis of thyroid cancer. However, accurate identification of nodule malignancy is a challenging task that can elude less-experienced clinicians. Recently, many computer-aided diagnosis (CAD) systems have been proposed to assist this process. However, most of them do not provide the reasoning of their classification process, which may jeopardize their credibility in practical use. To overcome this, we propose a novel deep learning (DL) framework called multi-attribute attention network (MAA-Net) that is designed to mimic the clinical diagnosis process. The proposed model learns to predict nodular attributes and infer their malignancy based on these clinically-relevant features. A multi-attention scheme is adopted to generate customized attention to improve each task and malignancy diagnosis. Furthermore, MAA-Net utilizes nodule delineations as nodules spatial prior guidance for the training rather than cropping the nodules with additional models or human interventions to prevent losing the context information. Validation experiments were performed on a large and challenging dataset containing 4554 patients. Results show that the proposed method outperformed other state-of-the-art methods and provides interpretable predictions that may better suit clinical needs.

Boundary-rendering network for breast lesion segmentation in ultrasound images.

Breast Ultrasound (BUS) has proven to be an effective tool for the early detection of cancer in the breast. A lesion segmentation provides identification of the boundary, shape, and location of the target, and serves as a crucial step toward accurate diagnosis. Despite recent efforts in developing machine learning algorithms to automate this process, problems remain due to the blurry or occluded edges and highly irregular nodule shapes. Existing methods often produce over-smooth or inaccurate results, failing the need of identifying detailed boundary structures which are of clinical interest. To overcome these challenges, we propose a novel boundary-rendering framework that explicitly highlights the importance of boundary for automated nodule segmentation in BUS images. It utilizes a boundary selection module to automatically focuses on the ambiguous boundary region and a graph convolutional-based boundary rendering module to exploit global contour information. Furthermore, the proposed framework embeds nodule classification via semantic segmentation and encourages co-learning across tasks. Validation experiments were performed on different BUS datasets to verify the robustness of the proposed method. Results show that the proposed method outperforms states-of-art segmentation approaches (Dice=0.854, IOU=0.919, HD=17.8) in nodule delineation, as well as obtains a higher classification accuracy than classical classification models.

The value of preoperative sentinel lymph node contrast-enhanced ultrasound for breast cancer: a large, multicenter trial.

OBJECTIVE: The study conducted a multicenter study in China to explore the learning curve of contrast enhanced ultrasound (CEUS) for sentinel lymph nodes (SLNs), the feasibility of using this technique for the localization of SLNs and lymphatic channels (LCs) and its diagnostic performance for lymph node metastasis. METHOD: Nine hundred two patients with early invasive breast cancer from six tertiary class hospitals in China were enrolled between December 2016 and December 2019. Each patient received general ultrasound scanning and SLN-CEUS before surgery. The locations and sizes of LCs and SLNs were marked on the body surface based on observations from SLN-CEUS. These body surface markers were then compared with intraoperative blue staining in terms of their locations. The first 40 patients from each center were included in determining the learning curve of SLN-CEUS across sites. The remaining patients were used to investigate the diagnostic efficacy of this technique in comparison with intraoperative blue staining and pathology respectively. RESULT: The ultrasound doctor can master SLN-CEUS after 25 cases, and the mean operating time is 22.5 min. The sensitivity, specificity, negative predictive value, and positive predictive value of SLN-CEUS in diagnosing lymph node metastases were 86.47, 89.81, 74.90, and 94.97% respectively. CONCLUSION: Ultrasound doctors can master SLN-CEUS with a suitable learning curve. SLN-CEUS is a feasible and useful approach to locate SLNs and LCs before surgery and it is helpful for diagnosing LN metastases.