now we are planning to update pretty old and narrow DICOM functionality
of our ultrasound modality and I'm hesitating over several subjects.
So, I would be grateful for any response concerning them:

1. As is well known, an usual image from US scanner consists of US
picture (or several) itself with some graphics and texts near and over
the picture. As far as I know, usually (or, at least, frequently) US
modalities simply and ugly "burn" graphics and texts into US picture
when saving in DICOM. Please, tell me, am I right that this approach is
still the state of the art in US DICOM? Do you know US scanners that do
it some other way? What approach can you recommend?

At the moment it seems like we should still "burn" because "overlays
see also

2. We have US Multi-frame IOD for dynamic images. It's good and native
for 2D-modes (B, B+Color Doppler etc) but not so native for M or D-like
modes (in which the US image indeed is a "long ribbon" that scroll or
sweep "through the screen"). In these modes only the very small part of
the image is really new at the next frame in comparison with the
previous one - so, in theory, there may be other approaches to save
such things effectively... Moreover, D-mode also is usually accompanied
by sound! Am I right that the standard does not deal with it
(concerning US) at the moment or have I missed something?

3. Some US scanners provide the possibility to load previously stored
images and measure some medical parameters (distances, areas, speeds
etc) on them in the same or similar way as user can measure on "fresh"
image. If we provide such functionality for our native DICOM-images and
if then user wants to save the image with new measurements, should it
be other US IOD image (in the same series and same study?) or, saying,
Presentation State?

4. Some scanners can even do some signal and image processing on loaded
images in the same way as during the scanning. If I want to do such
saving and loading in DICOM-compatible manner, is it a good approach to
use standard IOD extended with private tags for any info and data
needed for such processing?

Thanks in advance!