A Novel Approach to Nodule Feature Optimization on Thin Section Thoracic CT

Samala, Ravi K.; Moreno, Wilfrido; You, Yuncheng; Qian, Wei

doi:10.1016/j.acra.2008.10.009

Cited by 18 publications

(13 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although these studies illustrate that low-level image features can be used to distinguish between malignant and benign nodules, it is important to incorporate radiologists' knowledge into the process and to understand the relationship between the image features and radiologists' annotations. Such understanding can not only improve diagnosis of malignant lung nodules, but also simplify and accelerate the radiology interpretation process as suggested by Kahn et al [10] In the medical imaging area, efforts to find the relationship between image features and subjective or semantic ratings were spearheaded by Barb et al [5], Raicu et al [19], and Samala et al [21] . Barb developed a framework that manages visual content of lung pathologies.…”

Section: Related Workmentioning

confidence: 99%

An investigation into the relationship between semantic and content based similarity using LIDC

Kim

Dasovich

Bhaumik

et al. 2010

Proceedings of the International Conference on Multimedia Information Retrieval

View full text Add to dashboard Cite

There is considerable research in the field of content-based medical image retrieval; however, few of the current systems investigate the relationship between the radiologists' visual impression of image similarity and the computer calculated content-based similarity. Furthermore, those research studies that investigate these relationships analyze the visual similarity with respect to degree of malignancy without including specific characteristics that are important in the diagnosis process.The creation of the NIH/NCI Lung Image Database Consortium (LIDC) dataset offers the opportunity to perform the proposed research. Each nodule out of the 932 distinct nodules (larger than 3mm in diameter) was delineated and annotated by up to four radiologists using nine semantic characteristics that are important in the lung nodule interpretation process. Using the LIDC images, we propose to encode the radiologists' characteristic-based similarity and further discover if there is any relationship between this conceptual/characteristic-based similarity and the contentbased similarity for lung nodule interpretation.Our preliminary results show that it is a challenging problem to model the characteristic-based and content-based relationships for a broad category of lung nodules. A correlation of only 0.1 was obtained between the characteristicbased similarity and the predicted characteristic-based similarity using an artificial neural networked trained on four types of low-level image features (size, intensity, shape, and texture) calculated for 640 random pairs of nodules. Future research is necessary to investigate the appropriateness of the considered image features to model both the variation in the human interpretation of the lung nodules and the perceived characteristic-based similarity.

show abstract

Section: Related Workmentioning

confidence: 99%

An investigation into the relationship between semantic and content based similarity using LIDC

Kim

Dasovich

Bhaumik

et al. 2010

Proceedings of the International Conference on Multimedia Information Retrieval

View full text Add to dashboard Cite

show abstract

“…Samala et al [8] defined nine feature descriptors that describe the nodule characteristics that were used in assessments by radiologists. These descriptors are: 1. subtlety; 2. internal structure; 3. calcification; 4. sphericity; 5. margin; 6. lobulation; 7. speculation; 8. texture and 9. malignancy.…”

Section: Introductionmentioning

confidence: 99%

Toward Precise Pulmonary Nodule Descriptors for Nodule Type Classification

Farag

Elhabian

Graham

et al. 2010

Medical Image Computing and Computer-Assisted Intervention – MICCAI 2010

View full text Add to dashboard Cite

Abstract.A framework for nodule feature-based extraction is presented to classify lung nodules in low-dose CT slices (LDCT) into four categories: juxta, well-circumscribed, vascularized and pleural-tail, based on the extracted information. The Scale Invariant Feature Transform (SIFT) and an adaptation to Daugman's Iris Recognition algorithm are used for analysis. The SIFT descriptor results are projected to lower-dimensional subspaces using PCA and LDA. Complex Gabor wavelet nodule response obtained from an adopted Daugman Iris Recognition algorithm revealed improvements from the original Daugman binary iris code. This showed that binarized nodule responses (codes) are inadequate for classification since nodules lack texture concentration as seen in the iris, while the SIFT algorithm projected using PCA showed robustness and precision in classification.

show abstract

“…L(I, t) = log P(I It) + log P( t) (38) To fmd this log-likelihood function, one needs to estimate the conditional P(II!) and the unconditional P (f) image models, and identify their parameters.…”

Section: Modeling For Segmentationmentioning

confidence: 99%

“…[38] defined nine feature descriptors that describe the nodule characteristics that were used in assessments by radiologists. These descriptors are: 1) subtlety; 2) internal structure; 3) calcification; 4) sphericity; 5) margin; 6) lobulation; 7) speculation; 8) texture; and 9) malignancy.…”

Section: Pulmonary Nodule Definitionsmentioning

confidence: 99%

“…Low-dose computed tomography (LDCT) scans has lower radiation risk, hence became abundant in lung screening research. In the literature various approaches and methodologies can be found to model the lung nodules, exploiting geometric and texture attributes of the nodules (e.g., [34], [36][37][38]). In general, there are three key steps for precise candidate nodule detection: 1) a realistic model (template) of lung nodules; 2) an efficient detection methodology; and 3) ability to reduce or eliminate the false positives in the detected candidates.…”

Section: Detection Of Small Size Lung Nodulesmentioning

confidence: 99%

See 1 more Smart Citation

Modeling small objects under uncertainties : novel algorithms and applications.

Farag

View full text Add to dashboard Cite

A Novel Approach to Nodule Feature Optimization on Thin Section Thoracic CT

Cited by 18 publications

References 22 publications

An investigation into the relationship between semantic and content based similarity using LIDC

An investigation into the relationship between semantic and content based similarity using LIDC

Toward Precise Pulmonary Nodule Descriptors for Nodule Type Classification

Modeling small objects under uncertainties : novel algorithms and applications.

Contact Info

Product

Resources

About