▪ Abstract Image segmentation plays a crucial role in many medical-imaging applications, by automating or facilitating the delineation of anatomical structures and other regions of interest. We present a critical appraisal of the current status of semiautomated and automated methods for the segmentation of anatomical medical images. Terminology and important issues in image segmentation are first presented. Current segmentation approaches are then reviewed with an emphasis on the advantages and disadvantages of these methods for medical imaging applications. We conclude with a discussion on the future of image segmentation methods in biomedical research.

Current methods in medical image segmentation.

结节病易误诊,据王洪武等~([1])收集国内18篇关于此病误诊的文献,误诊率高达63.2%,当然有误诊就会有误治,如孙永昌等~([2])报道26例结节病在影像学检查诊断的第一印象中拟诊结核5例,其中就有2例完成规范的抗结核治疗,为此应引起临床对本病诊治的重视。

Sarcoidosis

Renal cell carcinoma (RCC) denotes cancer originated from the renal epithelium and accounts for >90% of cancers in the kidney. The disease encompasses >10 histological and molecular subtypes, of which clear cell RCC (ccRCC) is most common and accounts for most cancer-related deaths. Although somatic VHL mutations have been described for some time, more-recent cancer genomic studies have identified mutations in epigenetic regulatory genes and demonstrated marked intra-tumour heterogeneity, which could have prognostic, predictive and therapeutic relevance. Localized RCC can be successfully managed with surgery, whereas metastatic RCC is refractory to conventional chemotherapy. However, over the past decade, marked advances in the treatment of metastatic RCC have been made, with targeted agents including sorafenib, sunitinib, bevacizumab, pazopanib and axitinib, which inhibit vascular endothelial growth factor (VEGF) and its receptor (VEGFR), and everolimus and temsirolimus, which inhibit mechanistic target of rapamycin complex 1 (mTORC1), being approved. Since 2015, agents with additional targets aside from VEGFR have been approved, such as cabozantinib and lenvatinib; immunotherapies, such as nivolumab, have also been added to the armamentarium for metastatic RCC. Here, we provide an overview of the biology of RCC, with a focus on ccRCC, as well as updates to complement the current clinical guidelines and an outline of potential future directions for RCC research and therapy.

Renal cell carcinoma.

Sources, Effects and Risks of Ionizing Radiation

Intensity-based classification of MR images has proven problematic, even when advanced techniques are used. Intrascan and interscan intensity inhomogeneities are a common source of difficulty. While reported methods have had some success in correcting intrascan inhomogeneities, such methods require supervision for the individual scan. This paper describes a new method called adaptive segmentation that uses knowledge of tissue intensity properties and intensity inhomogeneities to correct and segment MR images. Use of the expectation-maximization (EM) algorithm leads to a method that allows for more accurate segmentation of tissue types as well as better visualization of magnetic resonance imaging (MRI) data, that has proven to be effective in a study that includes more than 1000 brain scans. Implementation and results are described for segmenting the brain in the following types of images: axial (dual-echo spin-echo), coronal [three dimensional Fourier transform (3-DFT) gradient-echo T1-weighted] all using a conventional head coil, and a sagittal section acquired using a surface coil. The accuracy of adaptive segmentation was found to be comparable with manual segmentation, and closer to manual segmentation than supervised multivariant classification while segmenting gray and white matter.

/pdf/adaptive-segmentation-of-mri-data-5ea2xd4wax.pdf

Adaptive segmentation of MRI data

Sixteen adrenal masses were identified with computed tomography (CT) in patients who had no clinical or biochemical evidence of adrenal pathology. In five patients, an adrenalectomy was performed, disclosing either an adenoma or nodular hyperplasia. Autopsy disclosed a cortical adenoma in one patient who died from unrelated causes. In the remaining 10 patients, follow-up scans 4--16 months later revealed no change in the size of the masses. A management plan of nonfunctioning adrenal masses discovered incidentally on CT is proposed.

Nonfunctioning adrenal masses: incidental discovery on computed tomography

Magnetic resonance (MR) imaging systems produce spatial distribution estimates of proton density, relaxation time, and flow, in a two dimensional matrix form that is analogous to that of the image data obtained from multispectral imaging satellites. Advanced NASA satellite image processing offers sophisticated multispectral analysis of MR images. Spin echo and inversion recovery pulse sequence images were entered in a digital format compatible with satellite images and accurately registered pixel by pixel. Signatures of each tissue class were automatically determined using both supervised and unsupervised classification. Overall tissue classification was obtained in the form of a theme map. In MR images of the brain, for example, the classes included CSF, gray matter, white matter, subcutaneous fat, muscle, and bone. These methods provide an efficient means of identifying subtle relationships in a multi-image MR study.

Multispectral analysis of magnetic resonance images.

Magnetic resonance (MR) images of 21 patients who had undergone radiation therapy were analyzed and compared with those of 15 patients who had untreated tumors. T2-weighted images (TR = 1,500 msec, TE = 90 msec) were most helpful in distinguishing recurrent tumor from radiation fibrosis. Radiation fibrosis, like muscle, usually remained low in signal intensity on T2-weighted images, while tumor demonstrated higher signal intensity. In no patient was the signal intensity of tumor the same or less than muscle on the T2-weighted images. However, relatively high signal intensity on T2-weighted images is not specific for tumor recurrence and may be seen in acute radiation pneumonitis, infection, hemorrhage, and even pulmonary radiation fibrosis.

Radiation fibrosis: differentiation from recurrent tumor by MR imaging: work in progress

Computed tomography (CT) provides a valuable new perspective in assessing abnormalities of the thorax. In the patient with a mediastinal mass or widening detected by plain chest radiography, a definitive diagnosis is sometimes possible which would not be obtainable by conventional noninvasive radiological techniques. In certain clinical situations, CT is a highly sensitive modality for examining the lungs. The proved and potential clinical uses of CT in solving diagnostic problems of the thorax are discussed.

Computed tomography of the thorax.

CT scans of 47 patients who had peripheral bronchogenic carcinoma contiguous to the pleural surface and who had undergone thoracotomy were retrospectively reviewed. The CT features of the primary neoplasm that were analyzed included the angle and amount of contact with the adjacent pleural surface, associated pleural thickening, fat plane between the tumor and chest wall, rib destruction, and chest wall mass. CT was of limited predictive value in separating those patients who had parietal pleural/chest wall involvement from those who did not. The combination of two or three CT findings (obtuse angle, greater than 3 cm contact with pleural surface, associated pleural thickening) resulted in a sensitivity of 87% and a specificity of 59%. The clinical symptom of focal chest pain, while not as sensitive (67%) as CT, was much more specific (94%) for parietal pleura/chest wall invasion.

Robert G. Levitt

Papers

Nonfunctioning adrenal masses: incidental discovery on computed tomography

Multispectral analysis of magnetic resonance images.

Radiation fibrosis: differentiation from recurrent tumor by MR imaging: work in progress

Computed tomography of the thorax.

Pleural and chest wall invasion in bronchogenic carcinoma: CT evaluation.