跳转到主内容

非常抱歉,我们不完全支持您的浏览器。如果您可以选择,请升级到较新版本或使用 Mozilla Firefox、Microsoft Edge、Google Chrome 或 Safari 14 或更高版本。如果您无法进行此操作且需要支持,请将您的反馈发送给我们。

全新设计的官网为您带来全新体验,期待您的反馈 在新的选项卡/窗口中打开

Elsevier
通过我们出版
Connect

Tackling bias by rethinking human anatomy education

2021年6月28日

Alison Bert, DMA

Group of people looking at tablet

Gender and racial bias is pervasive in human anatomy education, but how should we address it? 3D4Medical’s Chief Design Officer asks participants at Women in Tech event

Pictured above: For their Complete Anatomy education platform, 3D4Medical’s product designers are looking for ways to make their anatomy models more inclusive of race and gender. This image is from the Complete Anatomy video, which you can watch below. When you think of human anatomy, what do you visualize?

Many of us picture a male figure, perhaps influenced by classic images like Leonardo Da Vinci’s Vitruvian Man 在新的选项卡/窗口中打开, the illustrations in anatomy textbooks, or even the comical figure in Milton Bradley’s Operation game from the 1960s.

Side by side image of Vitruvian Male and Operation game

Irene Walsh used these examples in a slide to introduce her presentation at the Women in Technology World Series.

These were among the examples Irene Walsh 在新的选项卡/窗口中打开 used in her interactive workshop at the recent Women In Technology World Series Online Festival 在新的选项卡/窗口中打开. With levity, she introduced a topic that has serious implications for how medicine is learned and practiced.

In her role as Chief Design Officer of Elsevier’s 3D4Medical, Irene works with leading 3D artists, medical experts, developers and designers to bring human anatomy to life in Complete Anatomy — an educational platform that enables students to interact in-depth with the body systems they’re studying.

Image of Irene Walsh

Irene Walsh

Crucially, their work in evolving the platform involves getting input from users and potential users — a practice Irene extended to this workshop. The 60+ participants had an opportunity to contribute to user research by submitting answers to questions on an online form.

“Bias comes in many shapes and forms and often impacts product decisions unconsciously,” she said. “Working in the anatomy education space, there are many biases to evaluate and overcome.”

Irene pointed out a reality that is perhaps so obvious that many people have given it little thought — that for centuries, human anatomy has been presented from the perspective of the White European male. And that bias has far-reaching consequences.

As an example, Irene shared a finding from a new report 在新的选项卡/窗口中打开 showing that in the UK, Black women are four times more likely to die in childbirth.

In fact, gender bias in research has been linked with a wide variety of unfavorable outcomes for women. In her Gendered Innovations 在新的选项卡/窗口中打开 project at Stanford University, Prof Londa Schiebinger has compiled an extensive body of case studies showing the need to integrate sex and gender analysis into research. For instance, women in Britain are 50% likelier to be misdiagnosed following a heart attack because heart failure trials generally use male participants. And because the reference body for car design is male, women involved in collisions are nearly 50% more likely to be seriously injured.

“Normal is a relative term”

As the workshop progressed, the need to address bias became increasingly apparent, but the solutions less so. For example, what if you were given a choice to look at a female model? At what stage of the month should that female be in? As Irene noted:

Unlike the male, any snapshot of the female will capture a moment in their reproductive life: Are they ovulating? In their luteal phase? Pregnant? Menopausal? How should this be handled? 

Many challenges stem from the concept of “normal.”

A huge challenge for Product Management here is balancing the desire to emphasize that ‘normal’ is a relative term; there are many more than 8 billion versions of the human body if you count those who went before and those who are yet to come.

This reality, combined with the complexities of the human body and human identity, make it even harder to fulfill the goal of Complete Anatomy: “to give students a clear, concise, foundational understanding of how all the parts of the body fit together — without using categorization into arbitrary and unscientific groups.”

To come up with solutions, Irene asked participants a series of open-ended questions. For example:

  • What types of body/bodies do you think medical students should see when they first open their medical resource?

  • How do you think the female model could be best represented in anatomy learning?

Their responses will be used by the Complete Anatomy team to make their human anatomy models even more inclusive and useful to the students who learn from them.

Responses included:

  • "There should be a variety of bodies shown initially, male, female, intersex, a range of skin colours, and a range of ages."

  • "Male and female at a minimum but would be curious about showing transgender too; certainly different skin tones also represented"

  • "I am amazed with all my anatomy training I never noticed this. Definitely want the female not only shown as being pregnant or lying down."

  • "Present the female model across the different stages of life as most of the time the female body is not attempting pregnancy."

Acknowledging variety — and avoiding stereotypes

Image of a slide guiding principles for broader representation and skin pigment color wheel

Irene Walsh used this slide to show the the guiding principles for broader representation and the skin pigment color wheel “with no end and no beginning.”

Irene demonstrated an ongoing research project that will enable the main anatomical model to portray people from a broad spectrum of backgrounds. She considers that another crucial but overlooked area in anatomy education:

Are we best preparing our students for the richness of cultures and backgrounds they’re going to encounter in their work? Do they see themselves reflected in how we design and build our products? 

To avoid stereotypical classifications, they came up with four guiding principles:

  • Remove the idea of a default skin color.

  • Avoid labels for model types.

  • Allow users to select pigmentation from a color wheel rather than a set order like “the infamous color charts of the past.”

  • Decouple facial features and skin pigmentation, allowing the customer to choose their combination to work with.

These principles are illustrated in the following slide, which shows the breadth of variety that can be achieved. Instead of simply changing a model’s skin color, the user can choose from different facial features.

Image of anatomy models with varying skin pigmentation

In this option, the user can choose various skin pigments and facial features.

The idea here is to shake up the standard stereotypes and, although acknowledging that skin color is important in clinical study for detecting melanomas and other skin conditions, to show how superficial pigmentation is in terms of anatomy.

This would give educators the tools to better prepare the students for the real world, with a clear statement on the variety they’ll encounter, but equally at a foundational level, emphasize how our underlying anatomy is broadly the same. 

After explaining their approach and showing the imagery, Irene asked participants a question:

Would you feel represented in this new approach? 

While many participants responded positively, a few flagged an issue that could reflect an unconscious bias that can influence even the most astute product designers: the female 3D models appeared to have longer eye lashes than the male models, and slightly more color on their lips and cheeks, giving the appearance of wearing makeup. And one person asked, "Why are they all so young?"

Irene said the appearance of make-up may well have been an example of inherent bias resulting from the photos and medical imagery they used as reference material, presenting women in a more idealized way than their natural state. In fact, that feedback led the 3D artists to revise the aesthetic to a more natural look.

After the small group conversations, a participant wrote:

Some thoughts we had is whether you will include different ages, different weights as well, and people who are born as not clearly either of the binary sexes.

Closing comments

The discussion was robust. From the comments, it was clear that participants were engaged and liked the inclusive approach of showing work in progress and having difficult conversations. For example:

  • "This product is fascinating. Awesome work to you and your team."

  • "Super cool project, thank you for sharing!!"

  • "The software is amazing"

  • "Amazing session, interactive, insightful! Thank you so much!"

  • "I love that you actually took this as input for your project. Helpful for us! Helpful for you! And how diversity should be approached. Thanks so much :-)"