LO
Sonographer — tasks, one by one
The unit of analysis is the task, not the job title. Each one below carries its direction, whether the judgement rests on evidence or on platform inference, the reasoning, and what it does not establish.
Every task on this page#
Getting the standard views
Being augmented≈ Platform inferencePositioning the probe and the patient to capture the standard images an examination requires, and judging when each one is good enough.
AI guidance now tells the person holding the probe how to move it and captures the clip when it predicts the image is diagnostic. The FDA's 2020 decision on one such product rests on a study in which registered nurses with no scanning experience acquired limited ten-view heart ultrasounds, and describes its automatic capture as emulating the way a sonographer knows an image is good enough. For a sonographer that makes the routine views faster; for others it opens a limited scan to people who are not sonographers.
The FDA decision covers a limited set of adult heart views, and images must be interpreted by qualified professionals; it does not show how widely the tool is used or that it replaces a full examination.
Taking measurements
Being augmented≈ Platform inferenceMeasuring chambers, vessels, organs and the baby — sizes, flows and standard calculations — for the doctor's report.
Ultrasound systems increasingly propose standard measurements automatically once the right image is on screen. The sonographer checks that the machine measured the right thing on the right frame, and corrects it when the anatomy or the image quality misleads the software.
This rests on what ultrasound software can do, not on a measurement of how many measurements are now made automatically.
Difficult patients and unexpected findings
Still human-led≈ Platform inferenceGetting usable images through difficult anatomy, and noticing something the request did not ask about and extending the scan to show it.
Guidance software is built for a fixed set of views on a cooperative patient. When the anatomy is difficult, or when something unexpected appears on the screen, the sonographer decides what else to look at and how — judgement exercised with the probe in hand, while the patient is still on the table.
This rests on what current guidance software is designed and authorised to do, not on a measurement of how often scans are extended.
Caring for the person on the table
Still human-led≈ Platform inferenceExplaining the examination, positioning patients who are in pain, and handling intimate and emotional scans such as pregnancy and transvaginal examinations.
Ultrasound is done with a hand on the patient, often in pain or at an anxious moment — a pregnancy scan, a lump being checked. Keeping the person comfortable, and knowing what can and cannot be said before the doctor has reported, is part of the examination itself.
This rests on the nature of the work, not on a measurement of how patients' experience of ultrasound has changed.
Checking and teaching scans done by others
New task≈ Platform inferenceReviewing image quality from point-of-care scans done by nurses, doctors and other staff, and training them.
As guided ultrasound lets people who are not sonographers capture limited scans, someone who knows what a diagnostic image looks like has to check their images and teach them. The FDA decision itself shows the direction: the product was validated with nurses who had never scanned, and its warnings leave responsibility for image quality with the user.
No source yet measures how many sonographers now review or teach point-of-care scanning; this is an inference from the direction of the work.