Purpose
To reduce the inter- and intra- rater variability as well as time and effort, a method for computer-assisted delineation of hematoma is proposed. Delineation of hematoma is done for further ...automated analysis such as the volume of hematoma, anatomical location of hematoma, etc. for proper surgical planning.
Methods
Fuzzy-based intensifier was used as a pre-processing technique for enhancing the computed tomography (CT) volume. Autoencoder was trained to detect the CT slices with hematoma for initialization. Then active contour Chan–Vese model was used for automated delineation of hematoma from CT volume.
Results
The proposed algorithm was tested on 48 hemorrhagic patients. Two radiologists have independently segmented the hematoma manually from CT volume. The intersection of two volumes was used as ground-truth for comparison with the segmentation performed by the proposed method. The accuracy was determined by using similarity matrices. The result of sensitivity, positive predictive value, Jaccard index and Dice similarity index were calculated as 0.71 ± 0.12, 0.73 ± 0.18, 0.55 ± 0.14, and 0.70 ± 0.12 respectively.
Conclusions
A new approach for delineation of hematoma is proposed. The algorithm works well with the whole volume. Similarity indices of the proposed method are comparable with the existing state of art.
Computational analysis on altered micro-nano-textural attributes of the oral mucosa may provide precise diagnostic information about oral potentially malignant disorders (OPMDs) instead of an ...existing handful of qualitative reports. This study evaluated micro-nano-textural features of oral epithelium from scanning electron microscopic (SEM) images and the sub-epithelial connective tissue from light microscopic (LM) and atomic force microscopic (AFM) images for normal and OPMD (namely oral sub-mucous fibrosis, i.e., OSF). Objective textural descriptors, namely discrete wavelet transform, gray-level co-occurrence matrix (GLCM), and local binary pattern (LBP), were extracted and fed to standard classifiers. Best classification accuracy of 87.28 and 93.21%; sensitivity of 93 and 96%; specificity of 80 and 91% were achieved, respectively, for SEM and AFM. In the study groups, SEM analysis showed a significant (p < 0.01) variation for all the considered textural descriptors, while for AFM, a remarkable alteration (p < 0.01) was only found in GLCM and LBP. Interestingly, sub-epithelial collagen nanoscale and microscale textural information from AFM and LM images, respectively, were complementary, namely microlevel contrast was more in normal (0.251) than OSF (0.193), while nanolevel contrast was more in OSF (0.283) than normal (0.204). This work, thus, illustrated differential micro-nano-textural attributes for oral epithelium and sub-epithelium to distinguish OPMD precisely and may be contributory in early cancer diagnostics.
Image-based early detection for diabetic retinopathy (DR) needs value addition due to lack of well-defined disease-specific quantitative imaging biomarkers (QIBs) for neuroretinal degeneration and ...spectropathological information at the systemic level. Retinal neurodegeneration is an early event in the pathogenesis of DR. Therefore, development of an integrated assessment method for detecting neuroretinal degeneration using spectropathology and QIBs is necessary for the early diagnosis of DR.
The present work explored the efficacy of intensity and textural features extracted from optical coherence tomography (OCT) images after selecting a specific subset of features for the precise classification of retinal layers using variants of support vector machine (SVM). Fourier transform infrared (FTIR) spectroscopy and nuclear magnetic resonance (NMR) spectroscopy were also performed to confirm the spectropathological attributes of serum for further value addition to the OCT, fundoscopy, and fluorescein angiography (FA) findings. The serum metabolomic findings were also incorporated for characterizing retinal layer thickness alterations and vascular asymmetries.
Results suggested that OCT features could differentiate the retinal lesions indicating retinal neurodegeneration with high sensitivity and specificity. OCT, fundoscopy, and FA provided geometrical as well as optical features. NMR revealed elevated levels of ribitol, glycerophosphocholine, and uridine diphosphate
-acetyl glucosamine, while the FTIR of serum samples confirmed the higher expressions of lipids and β-sheet-containing proteins responsible for neoangiogenesis, vascular fragility, vascular asymmetry, and subsequent neuroretinal degeneration in DR.
Our data indicated that disease-specific spectropathological alterations could be the major phenomena behind the vascular attenuations observed through fundoscopy and FA, as well as the variations in the intensity and textural features observed in OCT images. Finally, we propose a model that uses spectropathology corroborated with specific QIBs for detecting neuroretinal degeneration in early diagnosis of DR.
Oral cancer is a leading malignancy and a rising concern in India. Early detection of the disease is essential at reducing mortality. In this paper, we propose a computer assisted method for ...diagnosis of oral pre-cancer/cancer using oral exfoliative cytology. A combination of features were extracted from expert delineated cells and nuclei collected from cytology of patients suffering from oral sub-mucous fibrosis, oral leukoplakia or oral squamous cell carcinoma and subject with no lesion. These features were used to train predictive machine learning models like support vector machine, k nearest neighbor, random forest, etc. These models were verified using validation data set. The verification experiments showed promising results with the random forest classifier having a test accuracy of 90%.
Lung cancer diagnosis is still a challenge because of the subjective differences among experts' opinions. While the patho-biological basis of lung lesion varies, it becomes imperative to take into ...consideration every diagnostic modality (macro and microscopic), but usually decision making follows a heuristic framework. A semi-automated fuzzy multi-criteria decision support system is proposed which considers the smeared knowledgebase (embedding heuristics) to understand the problem and adopts a systemic approach to understand lung cancer by malignant potentiality indexing (MPI) and score prediction. Our proposed technique is able to assist for acquisition of information from the experts, analysis of the information, detection of disease potentiality, diagnosis and prognosis through MPI and it may act as an input for patient's pathological assessment and treatment modality.
PDF