DIY brain anatomy: looking for transgender features (part 1)

This is my pre-hormone-taking adult brain:

In my post “the science of gender identity (part 2: brain anatomy)“, I describe evidence that certain regions of transsexual brains resemble in size the regions of their cisgender counterparts in the transsexuals’ gender identity.

So I would like to measure my brain region sizes to see how they stack up. Enter image recognition:

This is someone else’s brain, but it shows a visual representation of an image recognition procedure’s results. I discovered I can use the same procedure on my MRI (but without being able to generate the cool image).

I ran the procedure on nine separate images of mine and compared the results by brain region. Here are the Spearman R values for the comparisons:

These results are very consistent, telling me that the image recognition program is consistent. This gives me confidence in using the results in later analyses comparing my brain region sizes to that of a population of cisgender brain MRIs.

Code, Data, and Procedure

Data, code, and instructions necessary to produce these results are attached.

code_and_data

the science of gender identity (part 4: summary)

To prepare for a book I intend to write on the science of gender identity, I drafted the following three blog posts to collect my thoughts. They are highly technical; I need to recast the content for the layperson. I also assembled some of my own biological data to analyze.

The first post covers the very little we know about the genetics involved. It specifically examines polymorphisms in particular genes and tests their correlation to transsexualism.

The second post I think is the most compelling. And it has the best pictures! It investigates brain anatomy in transsexuals and how it differs from that of cisgendered individuals.

Finally, the third post covers some of the most recent psychological information I could get my hands on. There were two problems here: Since I’m not a psychologist I only understood the statistical arguments in the papers, and have very little access to psychological literature. Nonetheless I did my best. The most notable discussion in this post is the description of a study examining the stability of gender identity in very young children.



My Own Data

Testosterone

While no correlation between testosterone level and the male-to-female transgender experience has ever been established, it is interesting that my natural testosterone level is extremely low. (This was measured before I started blocking my testosterone with Spironolactone). Here is where I sit on the curve for natal males my age:

The “normal boundaries” are those that my HMO says are healthy. To produce the curve I extracted the mean and standard deviation from [1]. I asked this source for the raw data so I could produce the actual data distribution, rather than the normal approximation, but they did not respond.

Brain Anatomy

I have a brain MRI recorded before I started taking hormones. This is important because hormones can alter brain anatomy. I’m attempting to use the 3DSlicer program [2] to measure the sizes of my various brain regions using image recognition. My intent is to compare the measurements to a body of (sort of) age-matched female and male brain MRIs I downloaded from [3].

Right now I’m struggling with the image recognition for my particular MRI, but I’ll figure it out and report the results on this blog.

An example of what this effort looks like in 3DSlicer is:

Related Posts

the science of gender identity (part 1: genetics)

the science of gender identity (part 2: brain anatomy)

the science of gender identity (part 3: psychology)

References

  1. http://www.ncbi.nlm.nih.gov/pubmed/21697255 (supplemental data)
  2. http://www.slicer.org
  3. http://www.loni.usc.edu

the science of gender identity (part 2: brain anatomy)

This is the second post in a mult-part series surveying the current science of gender identity, particularly with regard to the transgendered population. In my previous post I discussed the proposed genetic associations and corresponding research. A future post, if I can find sufficient data, will address neuropsychology research related to the transgender experience.

Here I discuss studies exploring differences in brain anatomy between transsexuals and cisgendered controls. The analysis is biased toward the male-to-female transsexual population versus the female-to-male population due to the availability of research data, which is unfortunate. As more research becomes available I will remedy this imbalance.

The research I describe below clearly points to structural differences in the brain between transsexual and cisgendered individuals. This is observational data—It still doesn’t answer why the structural differences emerge in the first place, though a predominant theory suggests variation in sex hormone uptake in the brain during fetal development is a cause [5]. My previous post discusses how this hypothesis connects to genetic features.

A Bit About the Words I’m Using

I’d prefer to use the umbrella term “transgender” to label the study participants described below. However, “transgender” is too broad, as the research I describe focused on those who particularly want to or have modified their bodies to become a member of a different sex, which not all transgendered individuals want to do. Therefore I use the medical term for this population: “transsexuals”.

Increased Putamen Volume

A comparison of MRIs from 24 male-to-female transsexuals, along MRIs from with 30 male and 30 female age-matched controls is reported in [1]. The transsexual study participants had not started hormone replacement therapy, therefore the research avoided a major possible confounding effect due to the possibility of hormone treatment altering brain anatomy [4]. However, the study did not control for the sexual orientation of subjects and controls, which has also been associated with brain anatomy [2, 3] and could confound the analysis. This decision was made because the sexual orientation of the controls was unknown since their MRIs came from a database that did not record that information.

The research found that male control and transsexual gray matter volumes were similar for all regions of the brain under investigation except the putamen, which was significantly larger for the transsexual group. The transsexual left and right putamen volume distributions were closer to that of the female controls, as shown in the following boxplots taken directly from the paper [1]:

A different view, also taken directly from the paper, displays the regions of the brain having significant volume difference between the transsexual subjects and male controls at p < 0.001 (FDR-corrected). Only the right putamen (in red) appears at this significance level.

These results suggest that male-to-female transsexuals carry a “feminized” putamen, which may help explain their differing gender identity compared to the male controls.

Increased Cortical Thickness

The authors of the study I described in the last section performed an additional study looking at cortical thickness differences between male-to-female transsexuals and cisgendered male controls [6]. As in the previous study, the subjects had not started hormone replacement therapy, and the researchers did not control for sexual orientation.

The study examined MRIs from 24 transsexuals and 24 age matched controls, comparing cortex thickness at thousands of points along the cortical surfaces. The statistical tests for difference at each point were corrected for multiple comparisons using false discovery rate. A map showing the significantly different regions, taken directly from the paper, is shown below [6]:

The following cortical regions were identified as different between the two groups: frontal cortex, orbito-frontal cortex, central sulcus, perisylvian regions, paracentral gyrus, pre/post-central gyrus, parietal cortex, temporal cortex, precuneus, fusiform, lingual, and orbito-frontal gyrus.

These findings strengthen the argument that brain anatomical differences exist between male-to-female transsexuals and cisgendered males.

Sex-Atypical Hypothalamus Activation

The research described in this section evaluated the hypothalamus activation of 12 male-to-female transsexuals when smelling two steroids known to elicit sex-differentiated responses: 4,16-androstadien-3-one (AND) and estra-1,3,5(10),16-tetraen-3-ol (EST) [7]. Data from a similar study by the researchers with an unspecified number of heterosexual male and female controls was available for comparison. The male-to-female transsexuals in the study were all heterosexual with regard to their birth sex; this eliminates the confounding effect of sexual orientation.

Both the transsexuals and the female controls experienced hypothalamus activation by AND, while the male controls experienced hypothalamus activation by EST. The activation heatmap image below, taken directly from the paper [7], illustrates the similar response to AND for male-to-female transsexuals (MFTR) and female controls (HeW), versus the response by the male controls (HeM). Furthermore, this image shows how the response to EST is distinct between transsexuals and control males.

This analysis suggests that transsexual hypothalamus activation by these steroids is birth-sex atypical.

White Matter Microstructure

[8] is the first analysis I found concerning female-to-male transsexualism. Unfortunately, I could only read the abstract since I couldn’t find the full article for free, but here is a summary of the findings:

Fractional anisotropy (FA) was performed on 18 female-to-male transsexuals, 24 male controls, and 19 female controls. The controls were heterosexual. The transsexual subjects had yet not started hormone replacement therapy. The FA procedure evaluated the white matter fibers of the whole brain.

FA values for the right superior longitudinal fasciculus, the forceps minor, and the corticospinal tract were compared between the groups. The values for the female-to-male transsexuals more closely resembled those of the control males than the control females. This lends support for the existence of brain structure differences between female-to-male transsexuals and cisgendered females.

Brain Anatomy More Congruent With Gender Identity Than Biological Sex
Another study [9] considered seven female-to-male transsexuals and ten male-to-female transsexuals simultaneously, along with age matched controls (eleven cis-females and seven cis-males). This seems small for a study, but the authors cite a limited subject pool. The subjects were given MRIs and brain regions distinguishable by the gender identity variable and the interaction between the gender identity variable and biological sex variable were identified.

The researchers identified four brain regions where gray matter volume of the transsexual subjects were identical to that of the controls sharing the subjects’ gender identity, but different from the controls sharing the subjects’ biological sex. The gray matter volume was higher in the right middle and inferior occipital gyri, the fusiform, the lingual gyri, and the right inferior temporal gyrus for male-to-female transsexuals and cis-female controls. In contrast, the gray matter volume was greater in the left pre-and postcentral gyri, the left posterior cingulate, the calcarine gyrus, and the precuneus in female-to-male transsexuals and cis-male controls.

To limit confounding effects, all transsexual recruits for the study were homosexual (with regard to birth sex) and had not started hormone replacement therapy. Nothing is stated in the paper about the sexual orientation of the controls.

Post Mortem Studies

Two post mortem studies are of note, though they are limited by small sample sizes and the fact that the male-to-female transsexual subjects involved had been treated by estrogen, which may impact brain plasticity [10].

The first study [11] observed that the size of male-to-female transsexuals’ bed nucleus of the stria terminalis was more typical of cis-female size. Similarly, [12] reported that male-to-female transsexuals’ volume and neuronal density of the interstial nucleus of the anterior hypothalamus was more cis-female typical.

Related Posts

the science of gender identity (part 1: genetics)

the science of gender identity (part 3: psychology)

References

  1. http://www.ncbi.nlm.nih.gov/pubmed/19341803
  2. http://www.ncbi.nlm.nih.gov/pubmed/17975723
  3. http://www.pnas.org/content/early/2008/06/13/0801566105.abstract
  4. http://www.eje-online.org/content/155/suppl_1/S107.full.pdf+html
  5. http://www.ncbi.nlm.nih.gov/pubmed/12492297
  6. http://www.ncbi.nlm.nih.gov/pubmed/23724358
  7. http://cercor.oxfordjournals.org/content/18/8/1900.full.pdf+html
  8. http://www.sciencedirect.com/science/article/pii/S0022395610001585
  9. http://www.ncbi.nlm.nih.gov/pubmed/24391851
  10. http://www.eje-online.org/content/155/suppl_1/S107.full.pdf+html
  11. http://www.nature.com/nature/journal/v378/n6552/abs/378068a0.html
  12. http://brain.oxfordjournals.org/content/131/12/3132