Summary: Single slice regions of interest with no multi-slice context or interval/thickness information may need to be reported as area only, not volume. Explicit interval/thickness information can and should be encoded. Thickness should be distinguished from interval.
Long Version.
Given a Region of Interest (ROI), no matter how it is encoded (as contours or segmented pixels or whatever), one can compute its area, using the pixel spacing (size) information. If a single planar ROI (on one slice) is grouped with a bunch of siblings on contiguous slices, then one can produce a sum of the areas. And if one knows the (regular) spacing between the slices (reconstruction interval in CT/MR/PET parlance), one can compute a volume from the sum of the areas multiplied by the slice spacing. Often one does not treat the top and bottom slice specially, i.e., the ROI is regarded as occupying the entire slice interval. Alternatively, one could consider the top and bottom slices (or both slices) as only being partially occupied, and perhaps halve the contribution of the top and bottom slices.
The slice interval is distinct from the slice "thickness" (Slice Thickness (0018,0050)), since data may be acquired and reconstructed such that there is either a gap between slices, or slices overlap, and in such cases, using the thickness rather than the interval would not return a volume representative of the object represented by the ROI(s). The slice interval is rarely encoded explicitly, and even if it is, may be unreliable, so one should compute the interval from the distance along the normal to the common orientation (parallel slices) using the Image Position (Patient) origin offset and the Image Orientation (Patient) row and column vectors. The Spacing Between Slices (0018,0088) is only officially defined for the MR and NM objects, though one does see it in CT images occasionally. In the past, some vendors erroneously encoded the gap between slices rather than the distance between their centers in Spacing Between Slices (0018,0088), so be wary of it.
This all presupposes that one does indeed have sufficient spatial information about the ROI available, encoded in the appropriate attributes, which is the case for 2D contours defined relative to 3D slices (e.g., SR SCOORDS with referenced cross-sectional images), 3D contours (e.g., SR SCOORD3D or RT Structure Sets), and Segmentation objects encoded as image objects with plane orientation, position and spacing.
And it works nicely down to just two slices.
But what if one only has one lonely slice? Then there is no "interval" per se.
For 2D contours defined relative to 3D image slices one could consult the adjacent (unreferenced) image slices and deduce the slice interval and assume that was applicable to the contour too. But for 3D contours and segmentation objects that stand alone in 3D space, and may have no explicit reference to the images from which they were derived, if indeed there were any images and if indeed those images were not re-sampled during segmentation, then there may be no "interval" information available at all.
The RT Structure Set does handle this in the ROI Contour Module, by the provision of an (optional) Contour Slab Thickness (3006,0044) value, though it may interact with an the associated Contour Offset Vector (3006,0045) such that the plane of the coordinates is not the center of the slab. See PS 3.3 Section C.8.8.6.2.
The Segmentation object, by virtue of inclusion of the Pixel Measures Sequence (functional group macro), which defines the Pixel Spacing, also requires the presence of the Slice Thickness attribute, but only if Volumetric Properties (0008,9206) is VOLUME or SAMPLED. And wouldn't you know it, the Segmentation IOD does not require the presence of Volumetric Properties :( That said, it is possible to encode it, so ideally one should; the question arises as to what the "thickness" of a segmentation is, and whether one should slavishly copy the slice thickness from the source images that were segmented, or whether one should use the interval (computed if necessary), since arguably one is segmenting the volume, regardless of how it was sampled. We should probably consider whether or not to include Spacing Between Slices (0018,0088) in the Pixel Measures Sequence as well, and to refine their definitions to make this clear.
The SR SCOORD3D content item attributes do not include interval or thickness. That does not prevent one from encoding a numeric content item to associate with it, though no standard templates currently do. Either way, it would be desirable to standardize the convention. Codes are already defined in PS 3.16 for 112225, DCM, “Slice Thickness”) and (112226, DCM, “Spacing between slices”) (these are used in the Image Library entries for cross-sectional images in the CAD templates).
Anyhow, from a recipient's perspective, given no explicit information and no referenced images there is no other choice than to report only area. If an image is referenced, and its interval or thickness are available, then one may be tempted to use it, but if they are different, which should one use? Probably the interval, to be consistent with the general case of multiple slices.
From a sender's perspective, should one explicitly encode interval or thickness information in the RT Structure Set, SR SCOORD3D, and Segmentation objects, even though it is not required? This is probably a good move, especially for single slice ROIs, and should probably be something considered by the standard for inclusion as a CP.
David
Saturday, October 19, 2013
Monday, October 14, 2013
Binge and Purge ... Archive Forever, Re-compress or Discard ... PACS Lifecyle Management
Summary: Technical solutions and standards existing for implementing a hodge-podge of varied retention policies; teaching and research facilities should hesitate before purging or recompressing though; separating the decision making engine from the archive is desirable.
Long version:
As we continue to "binge" on imaging modalities that produce ever large quantities of data, such as MDCT, breast tomosynthesis and maybe one day whole slide imaging, the question of duration of storage becomes more pressing.
An Australian colleague recently circulated a link to a piece entitled "What should we do with old PACS images?", in which Kim Thomas from eHealth Insider magazine discusses whether or not to discard old images, and how. The article nicely summarizes the UK situation, and concludes with the usual VNA hyperbole, but fails to distinguish the differences in practice settings in which such questions arise.
In an operational environment that is focused only on immediate patient care, risk and cost minimization, and compliance with regulatory requirements, the primary questions are whether or not it is cheaper to retain, re-compress or delete studies that are no longer necessary, and whether or not the technology in use is capable of implementing it. In such environments, there is little if any consideration given to "secondary re-use" of such images, such as for research or teaching. Typically a freestanding ambulatory setting might be in such a category, the priorities being quality, cost and competitiveness.
An extreme case of "early discarding" arises in Australia where, as I understand it, the policy of some private practices (in the absence of any statutory requirement to the contrary) is to hand the images to the patient and discard the local digital copy promptly. Indeed, this no doubt made sense when the medium was radiographic (as opposed to printed) film.
In many jurisdictions though, there is some (non-zero) duration required by a local regulation specific to medical imaging, or a general regulation for retention of medical records that includes images. Such regulations define a length of time during which the record must be stored and made available. There may be a statutory requirement for each facility to have a written policy in place.
In the US, the HIPAA Privacy Rule does not include medical record retention requirements, and the rules are defined by the states, and vary (see for instance, the ONC summary of State Medical Record Laws). Though not regulatory in nature, the ACR–AAPM–SIIM Technical Standard For Electronic Practice of Medical Imaging requires a written policy and that digital imaging data management systems must provide storage capacity capable of complying with all facility, state, and federal regulations regarding medical record retention. The current policy of the ACR Council is described in Appendix E Ownership, Retention and Patient Access to Medical Records of the 2012-2012 Digest of Council Actions. This seems a bit outdated (and still refers to "magnetic tapes" !). Google did reveal a draft of an attempt to revise this, but I am not sure of the status of that, and I will investigate whether or not our Standards and Interoperability group can help with the technical details. I was interested though, to read that:
"The scope of the “discovery rules” in other states mean that records should conceivably be held indefinitely. Evidence of “fraud” could extend the statute of limitations indefinitely."
Beyond the minimum required, whatever that might be, in many settings there are good reasons to archive images for longer.
In an academic enterprise, the needs of teaching and research must be considered seriously, and the (relatively modest) cost of archiving everything forever must be weighed against the benefit of maintaining a durable longitudinal record in anticipation of secondary re-use.
I recall as a radiology registrar (resident in US-speak) spending many long hours in film archives digging out ancient films of exotic conditions, using lists of record numbers generated by queries for particular codes (which had been diligently recorded in the limited administrative information system of the day), for the purpose of preparing teaching content for various meetings and forums. These searches went back not just years but decades, if I remember correctly. This would not have been possible if older material had been discarded. Nowadays in a teaching hospital it is highly desirable that "good cases" be identified, flagged, de-identified and stored prospectively (e.g., using the IHE Teaching File and Clinical Trial Export (TCE) profile). But not everyone is that diligent, or has the necessary technology deployed, and there will remain many situations in which the value of a case is not recognized except in retrospect.
Retrospective research investigations have a place too. Despite the need to perform prospective randomized controlled trials there will always be a place for observational studies in radiology. Quite apart from clinical questions, there are technical questions to be answered too. For example, suppose one wanted to compare the performance of irreversible compression algorithms for a specific interpretation task (or to demonstrate non-inferiority compared to uncompressed images). To attain sufficient statistical power to detect the absence of a small but clinically significant difference in observer performance, a relatively large number of cases would be required. Obtaining these prospectively, or from multiple institutions, might be cost prohibitive, yet a sufficiently large local historical archive might render the problem tractable. The further the question strays from those that might be answered using existing public or sequestered large image collections (such as those available through the NBIA or TCIA or ADNI or CardiacAtlas), the more often this true.
Such questions also highlight the potential danger of using irreversible compression as a means of reducing storage costs for older images. Whilst such a strategy may or may not impinge upon the utility of the images for prior comparison or evidential purposes, they may render them useless for certain types of image processing research, such as CAD, and certainly so for research into compression itself.
Technologically speaking, as the eHI article reminds us, not all of the installed base of PACS have the ability to perform what is colloquially referred to as "life cycle management", especially if it is automated in some manner, based on some set of rules that implement configurable local policy. So, even if one decides that it is desirable to purge, one may need some technology refreshment to implement even a simple retention policy.
This might be as "easy" as upgrading one's PACS to a more recent version, or it might be one factor motivating a PACS replacement, or it might require some third party component, such as a VNA. One might even go so far as to separate the execution of the purging from the decision making about what to purge, using a separate "rules engine", coupled with a standard like IHE Image Object Change Management (IOCM) to communicate the purge decision (as I discussed in an old thread on Life Cycle Management in the UK Imaging Informatics Group). We added "Data Retention Policy Expired" as a KOS document title in DICOM CP 1152 specifically for this purpose.
One also needs a reliable source of data to drive the purging decision. Some parameters like the patient's age, visit dates, condition and types of procedure should be readily available locally; others may not, such as whether or not the patient has died. As I mentioned in that same UK thread, and has also been discussed in lifecycle, purging and deletion threads in the pacsadmin group, in the US we have the Social Security Administration's Death Master File available for this.
Since the necessary information to make the decision may not reside in the PACS or archive, but perhaps the HIS or EHR, separating the decision maker from the decision executor makes a lot of sense. Indeed, when you think about it, the entire medical record, not just the images, may need to be purged according to the same policy. So, it seems sensible to make the decision in one place and communicate it to all the places where information may be stored within an enterprise. This includes not only the EHR and radiology, but also the lab, histopathology, cardiology, and the visual 'ologies like ophthalmology, dermatology, etc. Whilst one day all databases, archives and caches may be centralized and consolidated throughout an enterprise (VNA panacea scenario), in the interim, a more loosely coupled solution is possible.
That said, my natural inclination as a researcher and a hoarder (with a 9 track tape drive and an 8" floppy drive in the attic, just in case) is to keep everything forever. Fortunately for the likes of me, disk is cheap, and even the power and HVAC required to maintain it are not really outrageously priced in the scheme of things. However, if you feel you really must purge, then there are solutions available, and a move towards using standards to implement them.
David
Long version:
As we continue to "binge" on imaging modalities that produce ever large quantities of data, such as MDCT, breast tomosynthesis and maybe one day whole slide imaging, the question of duration of storage becomes more pressing.
An Australian colleague recently circulated a link to a piece entitled "What should we do with old PACS images?", in which Kim Thomas from eHealth Insider magazine discusses whether or not to discard old images, and how. The article nicely summarizes the UK situation, and concludes with the usual VNA hyperbole, but fails to distinguish the differences in practice settings in which such questions arise.
In an operational environment that is focused only on immediate patient care, risk and cost minimization, and compliance with regulatory requirements, the primary questions are whether or not it is cheaper to retain, re-compress or delete studies that are no longer necessary, and whether or not the technology in use is capable of implementing it. In such environments, there is little if any consideration given to "secondary re-use" of such images, such as for research or teaching. Typically a freestanding ambulatory setting might be in such a category, the priorities being quality, cost and competitiveness.
An extreme case of "early discarding" arises in Australia where, as I understand it, the policy of some private practices (in the absence of any statutory requirement to the contrary) is to hand the images to the patient and discard the local digital copy promptly. Indeed, this no doubt made sense when the medium was radiographic (as opposed to printed) film.
In many jurisdictions though, there is some (non-zero) duration required by a local regulation specific to medical imaging, or a general regulation for retention of medical records that includes images. Such regulations define a length of time during which the record must be stored and made available. There may be a statutory requirement for each facility to have a written policy in place.
In the US, the HIPAA Privacy Rule does not include medical record retention requirements, and the rules are defined by the states, and vary (see for instance, the ONC summary of State Medical Record Laws). Though not regulatory in nature, the ACR–AAPM–SIIM Technical Standard For Electronic Practice of Medical Imaging requires a written policy and that digital imaging data management systems must provide storage capacity capable of complying with all facility, state, and federal regulations regarding medical record retention. The current policy of the ACR Council is described in Appendix E Ownership, Retention and Patient Access to Medical Records of the 2012-2012 Digest of Council Actions. This seems a bit outdated (and still refers to "magnetic tapes" !). Google did reveal a draft of an attempt to revise this, but I am not sure of the status of that, and I will investigate whether or not our Standards and Interoperability group can help with the technical details. I was interested though, to read that:
"The scope of the “discovery rules” in other states mean that records should conceivably be held indefinitely. Evidence of “fraud” could extend the statute of limitations indefinitely."
Beyond the minimum required, whatever that might be, in many settings there are good reasons to archive images for longer.
In an academic enterprise, the needs of teaching and research must be considered seriously, and the (relatively modest) cost of archiving everything forever must be weighed against the benefit of maintaining a durable longitudinal record in anticipation of secondary re-use.
I recall as a radiology registrar (resident in US-speak) spending many long hours in film archives digging out ancient films of exotic conditions, using lists of record numbers generated by queries for particular codes (which had been diligently recorded in the limited administrative information system of the day), for the purpose of preparing teaching content for various meetings and forums. These searches went back not just years but decades, if I remember correctly. This would not have been possible if older material had been discarded. Nowadays in a teaching hospital it is highly desirable that "good cases" be identified, flagged, de-identified and stored prospectively (e.g., using the IHE Teaching File and Clinical Trial Export (TCE) profile). But not everyone is that diligent, or has the necessary technology deployed, and there will remain many situations in which the value of a case is not recognized except in retrospect.
Retrospective research investigations have a place too. Despite the need to perform prospective randomized controlled trials there will always be a place for observational studies in radiology. Quite apart from clinical questions, there are technical questions to be answered too. For example, suppose one wanted to compare the performance of irreversible compression algorithms for a specific interpretation task (or to demonstrate non-inferiority compared to uncompressed images). To attain sufficient statistical power to detect the absence of a small but clinically significant difference in observer performance, a relatively large number of cases would be required. Obtaining these prospectively, or from multiple institutions, might be cost prohibitive, yet a sufficiently large local historical archive might render the problem tractable. The further the question strays from those that might be answered using existing public or sequestered large image collections (such as those available through the NBIA or TCIA or ADNI or CardiacAtlas), the more often this true.
Such questions also highlight the potential danger of using irreversible compression as a means of reducing storage costs for older images. Whilst such a strategy may or may not impinge upon the utility of the images for prior comparison or evidential purposes, they may render them useless for certain types of image processing research, such as CAD, and certainly so for research into compression itself.
Technologically speaking, as the eHI article reminds us, not all of the installed base of PACS have the ability to perform what is colloquially referred to as "life cycle management", especially if it is automated in some manner, based on some set of rules that implement configurable local policy. So, even if one decides that it is desirable to purge, one may need some technology refreshment to implement even a simple retention policy.
This might be as "easy" as upgrading one's PACS to a more recent version, or it might be one factor motivating a PACS replacement, or it might require some third party component, such as a VNA. One might even go so far as to separate the execution of the purging from the decision making about what to purge, using a separate "rules engine", coupled with a standard like IHE Image Object Change Management (IOCM) to communicate the purge decision (as I discussed in an old thread on Life Cycle Management in the UK Imaging Informatics Group). We added "Data Retention Policy Expired" as a KOS document title in DICOM CP 1152 specifically for this purpose.
One also needs a reliable source of data to drive the purging decision. Some parameters like the patient's age, visit dates, condition and types of procedure should be readily available locally; others may not, such as whether or not the patient has died. As I mentioned in that same UK thread, and has also been discussed in lifecycle, purging and deletion threads in the pacsadmin group, in the US we have the Social Security Administration's Death Master File available for this.
Since the necessary information to make the decision may not reside in the PACS or archive, but perhaps the HIS or EHR, separating the decision maker from the decision executor makes a lot of sense. Indeed, when you think about it, the entire medical record, not just the images, may need to be purged according to the same policy. So, it seems sensible to make the decision in one place and communicate it to all the places where information may be stored within an enterprise. This includes not only the EHR and radiology, but also the lab, histopathology, cardiology, and the visual 'ologies like ophthalmology, dermatology, etc. Whilst one day all databases, archives and caches may be centralized and consolidated throughout an enterprise (VNA panacea scenario), in the interim, a more loosely coupled solution is possible.
That said, my natural inclination as a researcher and a hoarder (with a 9 track tape drive and an 8" floppy drive in the attic, just in case) is to keep everything forever. Fortunately for the likes of me, disk is cheap, and even the power and HVAC required to maintain it are not really outrageously priced in the scheme of things. However, if you feel you really must purge, then there are solutions available, and a move towards using standards to implement them.
David
Subscribe to:
Posts (Atom)