Efficacy of texture analysis of ultrasonographic images in the differentiation of metastatic and non-metastatic cervical lymph nodes in patients with squamous cell carcinoma of the tongue

Kawashima, Y; Miyakoshi, Masaaki; Kawabata, Yoshihiro; Indo, Hiroko P.

doi:10.1016/j.oooo.2023.04.012

Cited by 7 publications

(4 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, ML applications have gained popularity in several fields, including but not limited to finance, e-commerce, security authentication, autonomous driving, and medicine [22]. Medical applications include efforts to discriminate between metastatic and non-metastatic disease [23], as well as the prediction of cancer treatment response [24][25][26] and likelihood of recurrence [27], just to name a few. Mining phenotypic information from images with radiomics analysis in conjunction with analytical ML algorithms presents a promising field of study to address countless clinical outcomes [28][29][30].…”

Section: Introductionmentioning

confidence: 99%

Deep Texture Analysis—Enhancing CT Radiomics Features for Prediction of Head and Neck Cancer Treatment Outcomes: A Machine Learning Approach

Safakish,

Sannachi,

Moslemi

et al. 2024

Radiation

View full text Add to dashboard Cite

(1) Background: Some cancer patients do not experience tumour shrinkage but are still at risk of experiencing unwanted treatment side effects. Radiomics refers to mining biomedical images to quantify textural characterization. When radiomics features are labelled with treatment response, retrospectively, they can train predictive machine learning (ML) models. (2) Methods: Radiomics features were determined from lymph node (LN) segmentations from treatment-planning CT scans of head and neck (H&N) cancer patients. Binary treatment outcomes (complete response versus partial or no response) and radiomics features for n = 71 patients were used to train support vector machine (SVM) and k-nearest neighbour (k-NN) classifier models with 1–7 features. A deep texture analysis (DTA) methodology was proposed and evaluated for second- and third-layer radiomics features, and models were evaluated based on common metrics (sensitivity (%Sn), specificity (%Sp), accuracy (%Acc), precision (%Prec), and balanced accuracy (%Bal Acc)). (3) Results: Models created with both classifiers were found to be able to predict treatment response, and the results suggest that the inclusion of deeper layer features enhanced model performance. The best model was a seven-feature multivariable k-NN model trained using features from three layers deep of texture features with %Sn = 74%, %Sp = 68%, %Acc = 72%, %Prec = 81%, %Bal Acc = 71% and with an area under the curve (AUC) the receiver operating characteristic (ROC) of 0.700. (4) Conclusions: H&N Cancer patient treatment-planning CT scans and LN segmentations contain phenotypic information regarding treatment response, and the proposed DTA methodology can improve model performance by enhancing feature sets and is worth consideration in future radiomics studies.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Texture Analysis—Enhancing CT Radiomics Features for Prediction of Head and Neck Cancer Treatment Outcomes: A Machine Learning Approach

Safakish,

Sannachi,

Moslemi

et al. 2024

Radiation

View full text Add to dashboard Cite

show abstract

“…PET/CT also involves exposure to ionizing radiation. Although Sun et al reported that the sensitivity and specificity of 18 FDG-PET/CT were 0.84 and 0.96, respectively, for detecting regional LN metastasis [12], PET/CT is still complementary to the conventional radiological investigations [13,14]. Recently, PET/MRI has been reported to have better diagnostic ability for LN metastasis than contrast-enhanced CT [1,15,16]; however, PET/MRI has not been widely used.…”

Section: Introductionmentioning

confidence: 99%

“…US is a simple, ionizing-radiation-free, non-invasive, and low-cost imaging modality; owing to its higher spatial resolution, it can facilitate better detection of intranodal architectural changes compared with CT and MRI [5][6][7]17]. Although US cannot evaluate deep LNs, such as the retropharyngeal nodes, it is the first-line modality for the intensive evaluation of relatively superficial LNs that appear enlarged on CT and/or MRI in patients with HNSCC [7,[18][19][20].…”

Section: Introductionmentioning

confidence: 99%

“…The US diagnosis of metastatic LNs is usually made using gray-scale (B-mode) US, which provides morphological information on the size, shape, borders, echotexture, and power, or color Doppler US, which provides vascular information [5,18]. Metastatic LNs are reportedly associated with absent hilar echoes, increase in the short-axis length, near circular form, and heterogeneous internal echoes in B-mode images; and compressed hilar flow, peripheral flow, and/or scattered parenchymal flow signals in power or color Doppler US [5,7,20].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Metastatic Lymph Node Detection on Ultrasound Images Using YOLOv7 in Patients with Head and Neck Squamous Cell Carcinoma

Eida,

Fukuda,

Katayama

et al. 2024

Cancers

View full text Add to dashboard Cite

Ultrasonography is the preferred modality for detailed evaluation of enlarged lymph nodes (LNs) identified on computed tomography and/or magnetic resonance imaging, owing to its high spatial resolution. However, the diagnostic performance of ultrasonography depends on the examiner’s expertise. To support the ultrasonographic diagnosis, we developed YOLOv7-based deep learning models for metastatic LN detection on ultrasonography and compared their detection performance with that of highly experienced radiologists and less experienced residents. We enrolled 462 B- and D-mode ultrasound images of 261 metastatic and 279 non-metastatic histopathologically confirmed LNs from 126 patients with head and neck squamous cell carcinoma. The YOLOv7-based B- and D-mode models were optimized using B- and D-mode training and validation images and their detection performance for metastatic LNs was evaluated using B- and D-mode testing images, respectively. The D-mode model’s performance was comparable to that of radiologists and superior to that of residents’ reading of D-mode images, whereas the B-mode model’s performance was higher than that of residents but lower than that of radiologists on B-mode images. Thus, YOLOv7-based B- and D-mode models can assist less experienced residents in ultrasonographic diagnoses. The D-mode model could raise the diagnostic performance of residents to the same level as experienced radiologists.

show abstract

Texture analysis of ultrasonography to differentiate metastatic from nonmetastatic cervical lymph nodes in mandibular gingival squamous cell carcinoma

Kawashima,

Abe,

Hagimoto

et al. 2024

Oral Science International

View full text Add to dashboard Cite

AimThis study aimed to distinguish between metastatic and nonmetastatic cervical lymph nodes (CLNs) in patients with mandibular gingival squamous cell carcinoma using texture analysis of lymph nodes on ultrasonography images.MethodsA total of 29 metastatic and 34 nonmetastatic CLNs were selected for this study. Fifty‐six texture characteristics were retrieved from the ultrasonography images using the LIFEx software.The Mann–Whitney U test was used to evaluate measurable differences in texture characteristics between metastatic and nonmetastatic CLNs. The ability of the surface features to distinguish between metastatic and nonmetastatic CLNs was illustrated using receiver operating characteristic curves. Youden's J statistic was used to determine the receiver operating characteristic curve cutoff values with maximal sensitivity and specificity.ResultsGray‐level nonuniformity, run length nonuniformity, coarseness, strength, and zone size nonuniformity showed the most notable contrast between the metastatic and nonmetastatic CLNs (p < 0.001). Zone size nonuniformity showed an area under the curve value of 0.827, sensitivity of 0.645, and specificity of 0.972 at the cutoff value of 1010.538.ConclusionQualitative characteristics may be the optimal surface features to distinguish between metastatic and nonmetastatic CLNs and predict CLN metastasis in patients with mandibular gingival squamous cell carcinoma.

show abstract

Efficacy of texture analysis of ultrasonographic images in the differentiation of metastatic and non-metastatic cervical lymph nodes in patients with squamous cell carcinoma of the tongue

Cited by 7 publications

References 48 publications

Deep Texture Analysis—Enhancing CT Radiomics Features for Prediction of Head and Neck Cancer Treatment Outcomes: A Machine Learning Approach

Deep Texture Analysis—Enhancing CT Radiomics Features for Prediction of Head and Neck Cancer Treatment Outcomes: A Machine Learning Approach

Metastatic Lymph Node Detection on Ultrasound Images Using YOLOv7 in Patients with Head and Neck Squamous Cell Carcinoma

Texture analysis of ultrasonography to differentiate metastatic from nonmetastatic cervical lymph nodes in mandibular gingival squamous cell carcinoma

Contact Info

Product

Resources

About