© 1999 Robert A. Freitas Jr. All Rights Reserved.

Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999

8.2 Human Somatography

Somatography is, quite simply, the "geography" or map of the anatomical spaces, as seen from the viewpoint of a microscopic traveler in those realms. The human body is a complex and fascinating place to visit. The nanomedical theater of operations in adult patients may range in size from an extremely small ~0.005 m³ (Lucia Zarate, 5.9 kg, in January 1883 at age 20) to an extremely large ~0.5 m³ (Robert Earl Hughes, 485 kg, in February 1958 at age 32),⁷³⁹ but averages ~0.06 m³ of navigable volume for the standard 70 kg adult male having 15% body fat, medium build and good health.⁸¹⁷ (The more typical tall but overweight late 20th century American male (6-ft, 220 lbs, 25% body fat) measures ~0.1 m³ in size.) Thus the volume normally available for nanorobot navigation in a single patient is ~10¹⁷ microns³.

The first fully digitized comprehensive three-dimensional map of the human body was compiled in the early 1990s as part of the National Library of Medicine's Visible Human Project.¹³⁰⁴ During this effort, a male and female cadaver were frozen in blocks of gel, sectioned into thousands of thin slices and digitally scanned in MRI (magnetic resonance imaging), CT (computerized X-ray tomography) and anatomical modes (optical photography), slice by slice.

The male data set, released in 1994, consists of axial MRI images of the head and neck taken at 4 mm intervals and longitudinal sections of the rest of the body also at 4 mm intervals; each MRI image is 256 x 256 pixels with a 12-bit grey scale per pixel. The CT data consists of 512 x 512 pixel axial CT scans of the entire body taken at 1 mm intervals, also with a 12-bit grey scale per pixel. The axial anatomical images are 2048 x 1216 pixels with a 24-bit color scale per pixel, representing ~60 megabits per image. The anatomical cross-sections are also at 1 mm intervals and coincide with the CT axial images. There are 1871 cross-sections for each mode (CT and anatomy) obtained from the male cadaver, a ~0.112 terabit data set for the anatomical images (~111 million micron³ per voxel).

The female data set, released in 1995, has the same characteristics as the male set except that the axial anatomical images were obtained at 0.33 mm intervals instead of 1.0 mm intervals, resulting in more than 5,000 anatomical images and a data set of ~0.336 terabits. Even though this map has a resolution of only 37 million micron³ per voxel, this is still sufficient to resolve all terminal veins and terminal arterial branches in the body and all major gross anatomical features, and thus provides a good start toward a human somatographic atlas.

A complete human somatographic static map to cellular resolution in theory requires ~20 micron increments, a ~100 terabit data set using 8-bit 8000 micron³ voxels assuming a fixed scanning geometry and ignoring the indexing tables. Recording all major structural details in the capillary terminal bed demands ~4 micron resolution, a ~10,000 terabit data set assuming 8-bit voxels. A 10,000 terabit data set may be stored on a ~26 bit/nm³ hydrofluorocarbon memory tape (Section 7.2.1.1) in a cubic volume of ~(72 micron)³ within an in vivo ~(100 micron)³ library nodule. Divided into ~10⁶ independent spools, a datum located anywhere on the 3300-kilometer total tape length could be accessed in ~10 seconds assuming a read speed of ~30 cm/sec (Section 7.2.6). A smaller 100 terabit library nodule requires only ~(16 micron)³ of tape, with ~1000 spools and similar access times as in the previous example. Topological or functional mapping may permit data compression by a factor of 10-100 (e.g., Sections 8.2.1.2 and 8.3.2) without increasing access time or seriously reducing map utility.

A 1 micron³ storage block that could conveniently be carried aboard an individual medical nanorobot can hold ~0.01 terabits (Section 7.2.6), enough memory to contain a three-dimensional map of the entire human body to ~430 micron resolution or a map of a 1-kg organ to ~100 micron resolution, using 8-bit voxels. The required resolution for a given application is very mission-dependent. Additionally, two types of map are likely to be of value. The first is a precise static map of some generic reference cadaver, as described above, that may provide general navigational guidance. The second is a detailed map of the individual patient, assembled by exploratory or "surveyor" nanorobots (Chapter 19) prior to the deployment of therapeutic nanorobots which would be given this information to allow very specific navigational guidance. Map stability is an important issue (Section 8.2.1.2; Chapter 19).

The remainder of this Section offers a quick guided tour of the larger navigable volumes inside the human body, along with some useful quantitative details. The different regions of the body appear to have enough structural and chemical dissimilarity to allow a nanorobot traversing these volumes to determine its position to at least ~mm accuracy, based solely on simple landmark recognition, chemonavigational cues, and bifurcation and topological information, even without resorting to the precision positional systems described in Section 8.3.

Last updated on 19 February 2003