We reported the investigation of the breast image viewing system, which has indicated many different kinds of digital mammography. The system is composed of Toshiba Medical Systems H/W:TDV-7000, S/W:STWS-005 and TFS-7000R. Four different CR systems and four different ultrasonography units were connected to this system. Following problems have rise to the surface; 1) the registration errors of the image type; although it must be registered as MG image, registered as CR image, 2) the output errors using a same examination UID for different patient or different examination, 3) the reading errors from the DICOM media, 4) the inharmonious display of the patients list using different characters for each manufactures, 5) the incorrect scale indication on the magnified photography, 6) the loss of the representation of the technologist's name, 7) the image quality insufficiency for reading diagnosis. Some problems are resolved and usual image interpretation of many kinds of diagnostic mammography can be performed now without any obstacles. The two direction four sheets mammography taken by PCM system, which has about 130 MB data for each one sheet can be developed within about nine seconds. Moreover, by the breast ultrasonographic images have been indicated on the monitor every two sheets, we can refine the breast scanning way. And very swift and functional multi-modality diagnosis has been realized in the conference using this breast image viewing system. Hereafter, more improvements of the system are expected.

View full abstract

At our institution, we have adopted digital mammography using a Senographe 2000D (GE, Co.) from the beginning of our screening program. Therefore we have not been able to directly compare the advantages and deficiencies of our adopted screen-film system with those of recently developed digital mammography based on our own hospital materials. However, our experience has allowed us to reach the following conclusions.
Advantages of monitor diagnosis: 1) Functions such as enlargement and changes in the gray level are useful for diagnosis. 2) Image information from extramammary structures such as the skin and subcutaneous tissue is readily obtainable from mammography images. 3) Searching for films is not necessary, and images can be visualized with a monitor. 4) Making a diagnosis is principally similar to the situation with film mammography.
Disadvantages of monitor diagnosis: 1) Because only a limited number of high-contrast resolution monitors can be made available in most consultation rooms and conference rooms, some other image demonstration instruments are necessary. 2) Full use of the monitor functions requires more reading time per case. 3) Concerning the shape of calcification, some changes in the image may occur as a result of enlargement or a change in condition, possibly leading to hesitation in the application of traditional judgment.
Care necessary for introduction of digital mammography: 1) We must try to read mammograms by optimizing the monitor settings. 2) Before purchase, we must critically determine the number of a high-contrast resolution monitors necessary, and the places where they should be installed, on the basis of clinical necessity and budget. 3) A fundamental necessity of digital mammography is reading ability, which requires adequate training in film mammogram reading.

View full abstract

The use of digital mammography has been rapidly popularized in Japan, but a questionnaire study has reported that soft-copy reading has been adopted in only 24% of all institutions. The potential advantages of digital mammography lie not in hard copy reading, but in soft-copy reading. Digital mammography has a wide dynamic range and therefore yields rich data by monitoring both the window level and the window width. When using a computer-aided diagnosis and comparing the present mammogram with previous ones, soft-copy diagnosis is more advantageous than hard-copy diagnosis. If we perform breast inspection and palpation with the aid of mammography reading, the quality of breast cancer screening will be improved.
In this paper, we present our practical method of mammography reading in detail, and stress how to apply the advantages of soft copy digital mammography to breast cancer screening.

View full abstract

We present our experience of the use of digital mammography, which was introduced at our institution in March, 2005. Images were obtained with a LORAD Selenia instrument with a direct conversion format, and reading was performed with a high-contrast BARCO's 5M monitor.
There are several advantages of digital mammography, such as shortening of the time from image acquisition until identification, ease in changing the conditions of image processing, and availability of comparison readings. Highly sharp images can be characteristically obtained by use of the direct conversion format, and specific identification of microcalcification can be made.
Digital mammography has several demerits, including an extremely expensive initial investment and the necessity of temperature control, although it has advantages in allowing the development of a system for remote image diagnosis and computer-aided diagnosis (CAD). Thus we can expect further developments in the future with the use of digital mammography.

View full abstract

[in Japanese]

View full abstract

Telemammography can eliminate the labor and time of film transportation and helps compensate for the uneven national distribution of mammographers and radiologists. However, about 90% of digital mammography machines in Japan utilize CR (computed radiography), and the majority of them necessitate hard copy diagnosis.
The purpose of this study was to clarify the usefulness of telemammography using soft-copy CR mammography in breast cancer screening. We interpreted the soft-copy CR mammograms of 9,439 screenees at Kochi Kenshin Clinic and transferred them to Kochi Medical School via optic fibers (provided by NTT Inc.), between July 2005 and March 2007. The recall rate was 6.8%, the cancer detection rate 0.36% and the positive predictive value 5.3%. These results were not inferior to those of the Kochi prefectural screening program using film screen (F/S) mammograms of 26,747 screenees during almost the same period (8.7%, 0.34% and 3.9%, respectively). Moreover, we were able to report the results of interpretations through the same line, thus enabling the screenees to know their own results without delay.
This study demonstrated that the effectiveness of telemammography using soft-copy CR is equivalent to that of F/S mammography, suggesting the feasibility of a telemammography network using CR mammography machines and facilities that are already in place.

View full abstract

A questionnaire survey about digital screening mammography in Japan was mailed to 1,900 institutions, eliciting a response rate of 44.2% from 49 university hospitals, 236 public or quasi-public hospitals, 363 private hospitals, 101 health or screening centers, and 72 clinics. We found that digital mammography techniques were widely used across Japan and that computed radiography (CR) systems were quite popular at 83%. A majority of the CR systems (74%) were Fuji CR systems, whereas the most common digital equipment with a flat-panel detector was the Senographe 2000D/DS. Hard copy reading accounted for more than half of all screening mammographic reading. Every current mammogram was compared with prior mammograms routinely in one-fourth of the cases, and on request only in half. The average glandular dose ranged from 0.7 to 3.54mGy (mean: 1.71mGy) for all of the digital equipment and was significantly lower with the Senographe 2000D/DS than with the CR systems (Tukey test: p<0.0001. We conclude that although daily quality control is performed in most institutions, further quality control for digital mammography should be developed, including that for monitors, and that this procedure should be included in the Japanese guidelines for quality assurance in mammography screening.

View full abstract

The purpose of this study was to learn more about digital screening mammography in Japan by conducting a questionnaire survey. We mailed the questionnaires to 1,900 institutes and achieved a response rate of 43.2%. We found that reasons for reading on hard copy included the added expense of a viewer system and familiarity with reading films hung on a light box, as well as the ability to read films simultaneously on-site. However, both hard copy and soft copy reading were employed for interpreting the screening mammograms. In soft copy reading, the time spent by the readers in interpreting the results varied more widely than for screen-film mammography. Most readers magnified the images and changed the window width and window level, using mainly 5-megapixel (MP) liquid crystal displays (LCDs). Although the use of 3-megapixel (MP) LCDs was not uncommon, 5MP LCDs were used in most cases when reading digital mammograms obtained with flat-panel detectors. In contrast, 3MP LCDs were used more frequently in reading computed radiography mammograms. The use of a computer-assisted detection (CAD) system for digital mammography was not prevalent in Japan, although a few readers used one to confirm their suspicions of a lesion. Readers who did not use the CAD system currently, however, anticipated doing so in the future.

View full abstract

If amorphous calcifications on stereotactic mammogram images become more readily identifiable visually by changing the parameters for multi-objective frequency processing, it is greatly helpful for examiners carrying out stereotactic mammotome biopsy. Seven cases with amorphous calcification on mammography underwent mammotome biopsy, and the stereotactic mammogram images were visually evaluated before and after local anesthesia. Among the parameters for multi-objective frequency processing, we tried to change both the multi-frequency enhancing type (MRT; R, P, or F) and degree of multi-frequency enhancement (MRE; 1.1 or 1.4) in order to obtain the best combination for good visualization, using the method of paired comparison. The best evaluation was obtained before anesthesia with parameters such as MRT: MRE= P: 1.4, and after anesthesia using MRT: MRE= P: 1.1. Previous parameters of MRT: MRE= R: 1.1 had been used.
In conclusion, this newly determined combination of parameters allowed better visualization of amorphous calcifications and provided more exact and rapid identification of disease foci. Thus, proper setting of parameters for multi-objective frequency processing appears to be important for better visualization of amorphous calcifications.

View full abstract

Stereotactic vacuum-assisted breast biopsy is a useful system for diagnosis of non-palpable breast lesions detected by microcalcification on mammography (MMG). We present our experience of repeat stereotactic vacuum-assisted breast biopsy (ST-VAB) for the diagnosis of non-palpable breast lesions. This study included 10 of 464 patients who underwent repeat biopsy using ST-VAB between September 1998 and September 2007 in our hospital. In these 10 patients, the lesions in 7 were histologically diagnosed as benign, 2 as ductal calcinoma in situ (DCIS), and one as invasive ductal carcinoma (IDC9 with small invasive components. Benign cases were diagnosed as category 4 or 5 on MMG, and their category did not change or improve after follow-up MMG from 6 to 12 months. Other lesions suspected to be malignant at first biopsy by ST-VAB were diagnosed accurately by repeated ST-VAB biopsy within a maximum period of 12 months. From these results, it is suggested that follow-up surgical excision is not necessary in cases diagnosed histologically as benign, and that patients with suspected malignancy require follow-up surgical excision.

View full abstract

Ultrasound breast cancer screening is useful for women with dense breast tissue or who are premenopausal. In typical ultrasound screening, a laboratory technician performs imaging, and a diagnostic radiologist assesses the images obtained. In the present study, the technician performing the imaging categorized the findings based on the same criteria a those used by the physician, and we then examined the disparities in interpretation between the technicians and the physicians.
Of 7,018 breasts in 3,514 women who underwent ultrasound screening (average age 44.0 years), 3,235 breasts with some type of finding were analyzed, and the rate of agreement in the categorization of findings was calculated. Categorization by the technician and physician matched for 3,083 breasts (95.3%). In breasts below category 3 where the categorization did not match (142 breasts), breast cancer was not noted. In breasts above category 3 where categorization did not match (10 breasts), breast cancer was present.
In primary interpretation to categorize findings based on images and lab reports, 4.67% of the breasts examined required further testing, but discussion of the findings with the technician caused the percentage of breasts requiring further testing to drop to 4.30%. In cases of mass-image-forming lesions, categorization tended to be lower, but in cases of non-mass-image-forming lesions the technician's categorization tended to be respected. Additionally, in cases of breast cancer the technician was unable to categorize instances of DCIS with hypoechoic areas alone or invasive cancer 8 mm or smaller in size with poorly defined margins as category 4 or higher.
In breast ultrasound screening, categorization of findings by the technician who performed the examination was helpful in conveying the opinion reached during the examination to the diagnostic radiologist. In addition, these two individuals' categorizations were readily comparable, and further clarified issues in question.

View full abstract

In 1990, Arakawa Borough, Tokyo, established a Cancer Prevention and Health Support Center, and in the next year as one of its main activities began to carry out breast cancer screening for women aged 30 years old or older, using inspection and palpation. After 1999, mammography screening was introduced for all first applicants aged 50 years or over. The results of the screening obtained during the 15-year period from 1991 through 2005 were reviewed and analyzed.
The annual application rate for screening was 15-19%, which was rather high for a large city. In relatively young applicants aged 30-49 years, for whom the screening method remained unchanged during the whole period, the breast cancer detection rate was 0.09%. The cancer detection rate in women aged 50 years or older was similar, at 0.10%, before the introduction of mammography, but thereafter the rate increased to 0.14%. This overall increase in the detection rate was thought to be due to the high detection rate of 0.74% obtained among the first applicants who underwent mammography screening.
It is suggested that mammography screening is necessary for all applicants aged 50 years or over.

View full abstract