Catálogo de publicaciones - libros

From 1994 until 2002, we performed 6956 ultrasonic screenings of the thyroid for all patients who received breast examination by ultrasonography. We discovered 14 thyroid cancers (0.32%) from 4327 cases with breast complaints. The incidence of thyroid cancer with breast cancer (0.25%) was three times higher than that of thyroid cancer without breast cancer (0.73%). It was concluded that ultrasonic screening of the thyroid was useful in patients with breast complaints.

Ductal carcinoma in situ (DCIS) is mostly diagnosed by mammography. The incidence is between 1% and 5% in countries without widespread use of mammography, but the incidence in mammography screening programs is between 8% and 25%. However, DCIS can also be diagnosed by galactography in patients with nipple discharge, by ultrasound, and sometimes also by MRI. These methods can tell precisely how big the cancer is and exactly where in the breast. However, to morphologically verify a suspicion of malignancy from any of these imaging methods, needle biopsy can be performed. By using needle biopsy, surgery can be planned more accurately as a curative measure instead of as a diagnostic biopsy. Fine-needle biopsy with thin needles for cytological diagnosis can be used successfully, especially in DCIS of the comedo type, but this technique is more operator dependent than core-needle biopsy and vacuum-assisted biopsy techniques where small pieces of tissue are sampled for histopathological analyses and diagnosis. There are no real complications for all these techniques. Patients diagnosed with DCIS have an excellent prognosis to survive without any local recurrences and general metastases.

Tissue elasticity imaging technology is expected to be a new modality for breast diagnosis, based on hardness as a tissue characteristic that is affected by tissue disease such as cancer. Different approaches of elasticity imaging have been investigated, and at present some are at the stage of developing a practical system. In clinical measurement, high-speed processing is required for real-time diagnosis, and freehand manipulation of the ultrasonic probe, such as in the usual ultrasonic diagnosis, is desirable for simple operation. Thus, we developed a tissue elasticity imaging system with freehand tissue compression based on a combined autocorrelation method. The method enables us to obtain tissue strain distribution as tissue elasticity with high speed and suppressing error due to lateral slip of the probe caused by freehand compression. The developed method was applied to breast disease measurements in vivo. Consequently, it is shown that the system can estimate the strain images in real time, and is effective not only in diagnosis of tissue hardness but also in determination of disease expansion area. It is also confirmed that the method is applicable to breast measurements in vivo.

For various soft tissues (e.g., liver, breast), we are developing the ultrasonic strain measurement-based mechanical properties (e.g., shear modulus, viscoshear modulus) reconstruction/imaging technique. To clarify the limitation of our quantitative reconstruction/imaging technique as a diagnostic tool for differentiating malignancies, together with improving the spatial resolution and dynamic range, we are collecting clinical reconstruction image data. Furthermore, we are applying our technique as a monitoring technique for the effectiveness of chemotherapy (anticancer drug, ethanol, etc.), thermal therapy (micro- and rf electromagnetic wave,HIFU, laser, etc.), and cryotherapy. Because soft tissues are deformed in 3-D space by externally situated quasistatic and/or low-frequency mechanical sources,multidimensional signal processing improves strain measurement accuracy and reduces inhomogeneity-dependent modulus reconstruction artifacts. These properties have been verified by us through simulations and phantom/human in vivo experiments. Briefly, here we discuss the limitations of low dimensional signal processing.Moreover, we exhibit the superiority on both differential diagnosis for these human in vivo malignancies and monitoring for these therapies of our quasi-real-time imaging (using conventional US equipment) to conventional B-mode imaging. Our technique is available as a clinical visualization technique both for diagnosis and treatment, and monitored mechanical properties data can also be effectively utilized as the measure for enhancing controlling therapy, such as the exposure energy, the foci, and the exposure interval. In the near future, suitable combination of various simple and low-invasive therapy techniques with our imaging technique will open up a new clinical style allowing diagnosis and subsequent immediate treatment. This approach should substantially reduce total medical expenses.

Detection of masses is much more difficult than that of microcalcifications (MCCs) because breast masses are part of tissues that may not be detected effectively by the techniques developed for detection of MCCs. In this chapter, we present a texture feature coding method (TFCM) to extract features that could characterize special properties of masses. It extracts gradient variations of gray level co-occurrence matrix as texture features. As a result, the TFCM is more sensitive to changes in texture. Three neural network architectures, backpropagation neural network, probabilistic neural network, and radial basis function neural network are used for mass detection with inputs provided by TFCM-extracted features. The experimental results show that our TFCM-based neural network approaches can achieve a detection rate of approximately 87% with a 10% false alarm rate.

Ultrasonic tissue elasticity imaging, which can visualize diseased tissues based on their stiffness, is a useful technique for breast cancer detection. In general, soft tissues including the breast have nonlinear elasticity and viscoelasticity referred to as high order mechanical properties. These properties make the conventional elasticity evaluations based on strain and Young’s modulus images difficult because these images vary depending on various conditions, in which high order mechanical properties cannot be neglected. For mechanical assessment independent of such conditions, high order mechanical properties must be assessed. Moreover, these properties also change in diseased tissues. Therefore, in this chapter, a method to assess high order mechanical properties is proposed with the aim of improvement of diagnostic ability. In actual breast assessment, cyclic loading and unloading were applied to the body surface by freehand manipulation of an ultrasonic probe; then, the nonlinear elasticity and viscoelastic hysteresis parameters were estimated and visualized based on the surface pressure measured with the pressure sensor, and the local strain distribution was estimated by the combined autocorrelation method. Nonlinear elasticity and hysteresis parameters, which can be estimated by the proposed method, clearly discriminated the breast tumor from the surrounding normal tissue.

From 1994 until 2002, we performed 6956 ultrasonic screenings of the thyroid for all patients who received breast examination by ultrasonography. We discovered 14 thyroid cancers (0.32%) from 4327 cases with breast complaints. The incidence of thyroid cancer with breast cancer (0.25%) was three times higher than that of thyroid cancer without breast cancer (0.73%). It was concluded that ultrasonic screening of the thyroid was useful in patients with breast complaints.

Depth to width ratio (D/W) is the only quantitive item in seven criteria used to evaluate breast masses by the Japanese Ultrasound Society. Recent technical improvement of ultrasound is remarkable, but the usefulness of the D/W has not fully been evaluated. We reviewed prospectively the D/W estimation of breast masses by using dedicated ultrasound units in ten hospitals that belong to the Japanese Breast and Thyroid Sonology and the Japanese Ultrasound Society. The materials included 163 regular invasive ductal carcinomas (CA) and 219 fibroadenomas (FA) that were selected from consecutive cases and which were smaller than 30 mm in diameter. The mean size of CA was 17 mm and that of FA was 14.5 mm. The D/W of CA was 0.76, significantly larger than that of FA, 0.57. The larger the size of the tumor, the smaller became the D/W in both CA and FA. The D/W of these two kinds of tumors was significantly different in cases with tumors larger than 5 mm in diameter. We also reviewed the D/W of three groups divided by the posterior echoes; that is, attenuating masses, no change, and accentuating. The D/W of the attenuating mass was larger than that of the other two groups. We reconfirmed that the D/W was useful in differentiating CA from FA.

To make a standard diagnosis, it is important to have a standard lexicon. Although many kinds of expression are possible for the same feature of a breast tumor, a technical term must be a word that is seldom confused with another different concept. Especially, a halo or boundary high echo must be a word that is used to express a malignant sign.

Many aspects of internal echoes of mass image-forming breast lesions have been discussed. These points, as well as other themes, should be documented or corrected through further investigation in wide clinical practice.