Mammography in Japan
Japanese: Nyubo X-sen kensa / mammography
You want to know whether a breast cancer can be found before you can feel it.
What it detects
- Microcalcifications — clustered calcium deposits behind 70–90% of ductal carcinoma in situ (DCIS) diagnoses made at screening
- Parenchymal abnormalities, especially architectural distortion, which become the more prominent sign after age 50
What the scan is looking for
A mammogram is an X-ray of the breast, and the finding it is best at picking up is not a lump. It is microcalcification — tiny clusters of calcium that appear white on the image. Most DCIS cannot be felt at all and is found only on imaging, and between 70% and 90% of DCIS diagnoses rest on spotting these clusters rather than on any palpable change.
Which sign matters most shifts with age. Microcalcifications are the most reliable indicator in women under 50. After 50, changes in the tissue itself — architectural distortion in particular — become the more visible warning.
That is why screening is offered at a stage when nothing can be felt. It is also why a normal mammogram is not the same as “no breast disease”: the test looks for a particular signature, not for everything.
How Japan runs it
Japan’s screening directive specifies the programme in more detail than many national schemes publish. Several of its provisions are worth knowing before booking anything here.
| Provision | What the directive says |
|---|---|
| Who it is offered to | Women aged 40 and over |
| Interval | Once every two years |
| Particularly encouraged | Ages 40–69 |
| Views taken | Mediolateral oblique (MLO) of both breasts. Women aged 40 to under 50 also receive a craniocaudal (CC) view |
| Reading | Two independent readers, at least one an experienced physician. Comparison with previous images is described as desirable |
Two points tend to surprise people.
The directive does not recommend visual and manual breast examination as part of screening. Where a facility performs it, the directive states it must accompany the X-ray rather than replace it.
And women in their thirties fall outside the programme. Rather than enrol them in routine screening, the directive asks that they receive guidance on breast awareness and on seeking specialist care promptly if something seems wrong.
What the accuracy figures actually say
The most frequently quoted recent benchmarks come from the US Breast Cancer Surveillance Consortium, which examined 458,175 digital breast tomosynthesis (DBT) screenings performed in community practice between 2011 and 2018: sensitivity 87.4% and specificity 92.2%.
DBT is a three-dimensional technique, while Japan’s municipal programme is built around conventional 2D mammography — so it is worth being precise about which of those numbers transfers. In that same study, sensitivity was similar between DBT and 2D digital mammography; it was specificity and the other measures that were higher for DBT. In other words, the sensitivity figure is a reasonable reference point for 2D screening, and the specificity figure is not.
These are US figures regardless. Japan’s programme differs in its reading requirements and in who is invited, and no equivalent published benchmark for the Japanese programme is cited here.
Limits of this examination
- Being called back is common, and usually not cancer. In a large US series of 458,175 tomosynthesis screenings, 8.3% were read as abnormal while cancer was found in 5.8 per 1,000 — roughly one call-back in twelve screenings, against about six cancers per thousand.
- Finding DCIS is not the same as finding a cancer that would have harmed you. DCIS covers a wide biological range: more than half of high-grade DCIS is thought to have the potential to become invasive within five years if left untreated, whereas low-grade DCIS carries that potential in 35–50% of cases and over a course as long as 40 years. Telling the two apart at the moment of detection is an open research question.
- A mammogram is an X-ray, and it is one test among several. Ultrasound and MRI answer different questions; a normal result means this particular signature was not seen, not that the breast was inspected in general.
- Accuracy figures belong to the equipment and reading process that produced them, and the numbers quoted here come from US community practice, not from Japan's programme.
Context for having this done in Japan
- Every screening image is read twice. The directive requires two independent readers, at least one of them an experienced physician, and describes comparison against the woman's previous images as desirable.
- The imaging protocol is specified nationally rather than left to each facility, down to which views are taken at which ages.
Descriptive only. We make no claim that outcomes here are better than anywhere else — that would need evidence we do not have.
Before you act on this
This page describes how an examination is defined and run in Japan. It does not tell you whether you should have one — that depends on your own history and is a conversation to have with a doctor. Where a facility offers this examination to visitors from abroad, the visit sits outside the public programme described above and is paid privately.
Sources
- がん予防重点健康教育及びがん検診実施のための指針 (Directive on cancer screening implementation) — Ministry of Health, Labour and Welfare, Japan (accessed 2026-07-26)
- Ductal Carcinoma In Situ of the Breast: An Update with Emphasis on Radiological and Morphological Features as Predictive Prognostic Factors — Cancers (Salvatorelli et al., 2020) (accessed 2026-07-26)
- National Performance Benchmarks for Screening Digital Breast Tomosynthesis: Update from the Breast Cancer Surveillance Consortium — Radiology (Lee et al., 2023) (accessed 2026-07-26)