https://d2ng68cf36kxsk.cloudfront.net/store/public/post/773/post_images/Slide218.JPG">

Not only is there a problem of accurate coding. There are two main classification schemes, ICD (the International Statistical Classification of Diseases and Related Health Problems)¹⁴² and SNOMED (Systematized Nomenclature of Medicine).¹⁴³ ICD exists in a succession of versions, older records in ICD-9, and newer records in ICD-10 and after that, ICD-11. IMO provides apps, accessible on computers and phones, for looking up the terminology associated with different medical conditions across ICD-9, ICD-10, ICD-11, SNOMED, and other specialized medical ontologies.

IMO also creates its own terms from synonyms that emerge in medical practice and are commonly found in medical records, and maps these to the standard ontologies. For instance, IMO has a term “abnormal excitement,” which maps to ICD version 9 code 799.29, “other symptoms involving emotional state.” Version 10 of ICD codes this R45.0, “nervousness.” SNOMED codes it 247006004, “Over-excitement,” or “Uncontrollable excitement.” At this point, medical classifications begin to run into another life-defining ontology, the Diagnostic and Statistical Manual of Mental Disorders.¹⁴⁴

In our grammar of multimodality, these are concepts,^§MS1.1.3 referencing things that exist or happen in more than one instance, and for that, begging classification. This is not to say that the material and the ideal align easily, the concept and its instances, the knowing and its being. Many of these concepts are matters of judgment and contention, where the ideal only roughly aligns with the material, where knowing only captures being as a matter of judgment. “Excitement,” “nervousness,” and “emotional states” traverse vast territories of human experience, and the point at which these become a medical condition as distinct from healthy life may at times become a contentious matter between patients, doctors, and insurance companies.

The official historians of ICD trace the origins of formal classification of medical conditions to registrations of the causes of death from Italy in the mid-fifteenth century and England in the mid-sixteenth. ICD had its beginnings in the International Statistical Congress which first met in Brussels in 1853, when the “CD” part of the acronym stood for “causes of death.” At the 1860 meeting in Paris, Florence Nightingale used death classification statistics to show the causes of hospital deaths and how they could be reduced.¹⁴⁵

The first version of the International List of Causes of Death was adopted at the Chicago congress of 1893.¹⁴⁶ Since then, ICD has gone through eleven major versions. The title “Causes of Death” was changed to “Classification of Diseases” in 1949 when the World Health Organization took responsibility for it. WHO now hosts periodical revision conferences and manages the revision process. ICD-10 was released in 1994, ICD-11 in 2018. ICD-11 expands the number of codes available to 55,000, up from the 14,400 in ICD-10.¹⁴⁷ Meanwhile, SNOMED, in development since 1965 and now controlled by a London-based not-for-profit, was created to describe a range of pathologies and clinical processes – 311,000 in total – not all of which are adequately captured in ICD.¹⁴⁸

ICD is arranged in what we have called “supermarket order,”^§MS0d where objects for practical purposes are put beside each other, only in one place, but with other things that are meaningfully associated. The practical process here is in the nature of written text – an essentially visual and spatial arrangement, where things roughly alike need to be placed near each other. Text in this respect is much more like image and space than it is like speech. There is nothing linear or temporal about the ICD text, as there is in the audio processes of speech.

Also, nobody could conceivably remember or be able to speak more than a few of 55,000 or 311,000 things, which is why IMO created the look-up app as a textual prosthesis for medical professionals. (Here again, “language” shows itself to be a singularly unhelpful concept, because text and speech are so practically unlike each other.)

ICD-10 is a carefully ordered classification scheme, in sections, marked auspiciously with roman numerals. It has sections on various bodily systems: IX “circulatory,” X “respiratory,” XI “digestive,” IV “endocrine.” Then there are some strange system conjunctions: XIII the “musculoskeletal system,” and XIV “genitourinary system” – muscles are very different from bones and reproduction very different from pissing, but in the body, these things work together or are near each other.

Then there are ragbags of other things, some internal to bodies, some external, but all objects of very different orders: organs or parts of the anatomy such as VIII “eyes,” VII “ears,” XII “skin”; acquired conditions such as I “infectious diseases” and II “neoplasms” (cancers); external effects such as XIX “injuries”; conditions that may have been inherited in the form of XVII “congenital malformations”; conditions that may not even be medical, at least in their origins, but which might now be classified as V “mental or behavioral disorders”; and stuff that happens in XXI “contact with health systems.”¹⁴⁹ Cross-classification clarifications are offered in the form of inclusions and exclusions. XVI “Perinatal conditions,” we are told, includes conditions whose origins are in pregnancy even though the baby dies later, but they exclude congenital malformations.¹⁵⁰

For all its agonizing order, and after a century and a half of institutional agonizing about its ordering, ICD still has the appearance of a ramshackle list. This is not because the ideal and our medical thinking is flawed, but because the material is varied and complex. It is as ramshackle as the particularities and relations of life itself. And it is as fallible as the politics of the construction of the medical self, where old maladies such as homosexuality are no longer that, and new conditions appear, such as post-traumatic stress disorder. “Classification systems,” say Bowker and Star, “simultaneously represent the world ‘out there,’ the organizational context of their application and the political and social roots of that context.”¹⁵¹

Then there is the frequently appearing but nevertheless disquieting notion of “other” – “other infectious diseases,” “unspecified mental disorder,” “neoplasms of uncertain or unknown behavior,” “other ill-defined and unspecified causes of mortality,” “provisional assignment of new diseases of uncertain etiology or emergency use,” to mention just a few places in ICD-10.¹⁵² If an ontology is to encompass all possibility in a domain, it has to countenance as-yet or in-the-moment unknown possibilities.

“Other” is a sleight of hand, where an ontology extends its range beyond the finite, anticipating the inevitable excesses of both the ideal and the material. While admitting its momentary finitude, the schema absorbs its certain failure to address everything that is conceivably addressable.

---

¹⁴² www.who.int/classifications/icd/en/

¹⁴³ www.snomed.org/

¹⁴⁴ www.psychiatry.org/psychiatrists/practice/dsm

¹⁴⁵ Mark Bostridge, 2008, Florence Nightingale: The Woman and Her Legend, London, UK: Viking.

¹⁴⁶ Iwao M. Moriyama, Ruth M. Loy, and Alastair H.T. Robb-Smith, 2011, History of the Statistical Classification of Diseases and Causes of Death, Hyattsville, MD: National Center for Health Statistics, pp. 1, 11–12.

¹⁴⁷ www.medscape.com/viewarticle/898202

¹⁴⁸ www.snomed.org/snomed-ct/what-is-snomed-ct/history-of-snomed-ct

¹⁴⁹ http://apps.who.int/classifications/icd10/browse/2016/en

¹⁵⁰ http://apps.who.int/classifications/icd10/browse/2016/en#/XVI

¹⁵¹ Geoffrey C. Bowker and Susan Leigh Star, 2000, Sorting Things Out: Classification and Its Consequences, Cambridge, MA: MIT Press, p. 61.

¹⁵² http://apps.who.int/classifications/icd10/browse/2016/en: B99, F99, D37–48, R99, U00–49.

---

Cope, Bill, and Mary Kalantzis. 2020. Making Sense: Reference, Agency and Structure in a Grammar of Multimodal Meaning. Cambridge UK: Cambridge University Press, pp. 303–307. https://doi.org/10.1017/9781316459645

Intelligent Medical Objects §MS3.2a