Custom Fit Bras
ASIM utilizes generative AI technology to provide a custom-fit bra service tailored to diverse body shapes and unique needs. This service disrupts traditional bra shopping by involving users directly in the design process, ensuring each piece is as unique as the individual wearing it.
ASIM redefines all experiences around bra shopping and wearing experiences.
How It Works:
Data Entry and Collection
Through ASIM’s user-friendly digital app, users engage in an interactive process where they can submit breast and data, as well as preferences and needs, through many offered methods.Data Analysis: Education and Insights
ASIM educates users about their bodies. By analyzing and tracking changes over time, it provides valuable insights, helping users understand their physical evolution and preferences.Data Actualization: Custom Creations
With precision and care, ASIM translates this data into perfectly tailored bras. Users can virtually try on their designs, make adjustments, and finalize their purchase, ensuring a fit that truly supports and enhances their comfort and confidence.
CHALLENGEPeople and breasts are complex.
Breasts hold multifaceted significance beyond just their anatomical functions, encompassing associations with motherhood, sexuality, beauty, gender, identity, and health. Yet, societal standards often impose a narrow view of how breasts should appear, failing to accommodate the natural diversity in breast size, shape, and the changes they undergo through life stages such as puberty, pregnancy, and aging. These standards also can clash with individual preferences in bra fit and style.
But bras, being designed with certain assumptions, push normative standards, instead of supporting the individual needs of users.
The design of traditional bras is built on these societal assumptions and does not reflect the reality of many wearers.
For example, the industry standardizes sizes based on a symmetrical, uniform model that ignores significant variations among individuals, such as those with unilateral mastectomies or naturally asymmetric breasts.
Furthermore, most sizing systems use a limited set of measurements, focusing primarily on rib and cup sizes, which do not account for the full diversity of body and breast shapes. This leads to isolating shopping experiences and physical discomfort for many, highlighting a clear disconnect between current market offerings and the actual needs of users.
OBJECTIVEComplexity needs to be embraced to ensure everyone feels their best, by helping them to find their ‘fit’ in every sense of the word.
To accomplish this, we set out to:
Design a digital product that allows users to submit data for custom bra creation and provides educational insights about their breasts over time.
Ensure the digital experience is inclusive and supports privacy, accuracy, and convenience.
Conduct market research and design a business model that supports the service.
Collaborate with experts to develop a generative algorithm for creating supportive bras based on user data.
Questions guiding research, concept testing and prototype development.
User Questions
How might we design for different degrees and types of anisomastia?
How might a series of custom fit products support pubescence, as well as gender transition?
How might bras support users through cancer-related needs such as tumor-related pain, swelling, and post-mastectomy prostheses?
Product Questions
How can the traditions of bra design be honored, and challenged, in new bra designs?
How might we account for variations in body shapes, sizes, and individual needs with new methods of measuring and bra fit analysis?
How might a bra with a diverse feature set consider the needs of cancer patients, child rearing and nursing individuals, gender transitioning individuals , and bodies undergoing change?
Technology Perception Questions
How might we support healthy body images while measuring anatomy and make necessary data capture less onerous on users?
How might we reassure users of data security, and implement and communicate privacy-preserving functionality?
RESEARCHUser research was approached with three methodologies.
A user survey to determine how prevalent breast asymmetry and bra discomfort is amongst a spectrum of individuals.
Expert interviews to understand the relationship between breast and body anatomy, bra construction, and manufacturing.
In-depth user interviews with a diverse range of users to understand their current experiences and expectations.
User SurveyA large problem, the survey uncovered that nearly half of individuals have asymmetric breasts, alongside universal bra discomfort.
Our survey of 125 individuals revealed widespread discomfort and dissatisfaction with current bra designs, highlighting issues particularly related to asymmetry and overall fit. Notably, nearly half of the participants reported natural breast asymmetry. Both groups, those with self-reported symmetrical and asymmetrical breasts, experienced similar bra problems. Common issues such as shoulder straps digging in, improper cup fill, and underwire discomfort were prevalent, underscoring fundamental flaws in design and sizing. Furthermore, only 35% of respondents felt moderately to extremely satisfied with their current bras, signaling a significant opportunity for improvement in the market.
Survey conducted in April 2022
n = 125 participants:
Ages: <15 = 1%, 16-30 = 70.5%, 31-45 = 11%, 46-60 = 7%, >61 = 10.5%
Genders, identified as: 91% = female, 1% = male, 6% = non-binary, 2% = transgender
Expert InterviewsTo design better, more inclusive bras, new ways for measuring, communicating and translating breast data into bras patterns are needed.
Experts were consulted to provide insights into breast anatomy, bra construction, and the manufacturing process. Specifically, they advised on the most effective data to collect, measurement techniques, and how to visually communicate this data. When exploring the types of data and their interrelationships, we adopted a holistic and iterative methodology. This approach ensures continuous feedback and refinement across all phases, facilitating a dynamic data flow that allows for constant adjustments in measurement, visualization, and bra design integration. By doing so, we ensure our designs are truly responsive to diverse body types, enhancing functionality and comfort.
User InterviewsAnd to drive acceptance and enhance the impact of custom-fit bra technology, it is crucial to understand user needs and expectations.
The interviews consisted of two sections.
Ethnographic Research: We explored their unique challenges and needs concerning bra shopping and wearing, and their perspectives on body confidence and image.
Concept Testing: We assessed their attitudes towards generatively designed, tailored-fit bras and discussed their preferences for various data submission methods.
12 individuals who have experienced greater degrees of changes or problems with their breasts were interviewed. This group included survivors of breast cancer and others who have undergone major surgeries, ensuring a comprehensive understanding of unique bra needs. The interviews were structured to capture detailed personal experiences and included participants such as:
A 27-year-old with a preventative bilateral mastectomy and direct-to-implant reconstruction due to the BRCA1 gene.
A 65-year-old who underwent a lumpectomy, unilateral mastectomy, and is currently preparing for implant reconstruction.
A 69-year-old with a unilateral mastectomy without subsequent reconstruction.
A 75-year-old who had a bilateral radical mastectomy followed by flap reconstruction.
A card sort of various images, methods, and technologies was used in the interviews to foster discussion, capture ideas, and gauge responses to different data collection concepts.
INSIGHTSBody-image issues are provoked across experiences.
Many users have felt uncomfortable in breast-related retail and medical settings, affected by non-inclusive sizing, and insensitive depictions of their bodies, triggering body image and self-confidence issues.
“I was 3D scanned by my plastic surgeon, it was very clinical... The imaging made me feel worse about my body than looking in the mirror. I didn’t see me, just a model of someone else’s body.”
Size, shapes, and styles doesn’t currently support all users.
Post-surgery users, struggle to find bras with necessary features like prostheses pockets, cups for adjusted nipple placements and varied intermammary clefts, and reinforced side support, underscoring the need for tailored designs.
“When I lay on my back, my implants go into my armpits, causing me numbness in my arms. It’s hard to find that a bra to sleep in that supports them.”
There is a gap in breast-health and bra fitting education.
Interviewees reported a lack of reliable information on breast health and bra fitting outside of medical consultations, with inconsistent advice and fitting experiences across retail experiences adding to the confusion.
“It’s very discouraging and awkward to try stuff on in the store, since they often don’t fit or have my size. So, I usually shop online.”
Data privacy is a top priority.
Despite recognizing the benefits of custom-fit bra services, interviewees, considering the privacy of breasts, expressed significant hesitation and insisted that their data be deleted if they were to use the service.
“I don’t really care but some people might wonder where those scans are going. As a shopper, I would just want to make sure the scans were deleted after.”
Precision expectations highly influence decision-making.
Users prioritize accuracy, demanding precise methods and clear evidence of the effectiveness in both data collection and anticipated bra fit. They’re even willing to use a less preferred technique if it promises greater accuracy.
“A mold would be the most accurate, but augmented reality is probably more realistic if it’s accurate and you guys can portray that that to get trust there.”
It needs to be easy - easier than just settling an ill-fitting bra.
Perceived ease-of-use influences the selection of a data collection methods and willingness to use custom-fit bra services. Users weigh convenience against expected accuracy.
“Even if it promises a super accurate fit, but it wasn’t convenient, I might not go for it, because it can just be emotionally and physically exhausting.”
It can’t be a one-size-fits-all service model.
Interviewees had different perceptions and attitudes towards the proposed data collection methods, driven by their own needs, comfort levels, and experiences, underscoring the importance in having a flexible service model.
DATA AND DESIGN PRINCIPLES:Research and insights are tied together to create a custom-fit bra service that helps users feel great, without compromise.
Navigating the complex web of associations, assumptions, and sizing challenges clearly illustrates why many endure discomfort.
To address this, we have defined specific types of data to be collected and utilized, and adhere to key design principles throughout the service design and product development.
User-Centric Data
Physical inputs
Placement and angle of breasts on chest
Shoulder, back, and rid-cage dimensions
Weight and density of breast
Elasticity of breast skin tissue
Degree of breast symmetry
Stage of breast ptosis
Nipple placement
Breast sensitivity
Shape of breast
Perceptional and preferential inputs
User’s degree of self-confidence
User’s body relationship
(e.g., post-nursing, gender confirmation)Bra style preferences
(e.g., balconette, plunge, sports)Desired appearance
(e.g., symmetry, perceived size)Material preferences, color, and padding level
Research and interviews highlight a strong interest for customized bra solutions tailored to individual preferences. While participants varied in their experience and technology preferences, they universally valued:
Privacy: Clear communication on data use and management.
Accuracy: Results that meet expectations.
Convenience: An efficient and straightforward process.
Sensitivity: Inclusive, accessible, and a safe-space.
Design Principles
DESIGNWith a better understanding of the users, required data, and priority design principles, I began designing how those features would be realized. ASIM’s three main capabilities include -
Data Entry and Collection
Data Analysis: Education and Insights
Data Actualization: Custom Creations
The design process included many methods, though not all shown, they included, iterative design sessions, user task flows, wireframes, style guide, design library, Figma prototypes and a service blueprint.
Design GuideThe brand identity was designed to comfort and resonate with users through fluidity, transparency, and sensitivity.
The Design Guide was developed to ensure that every aspect of our app and digital products not only adheres to our brand's core values but also enhances user interaction through consistency, efficiency, and scalability.
By establishing a strong and cohesive visual language, we facilitate a seamless and engaging user experience that reflects our commitment to fluidity, transparency, and sensitivity. This approach not only strengthens our brand identity but also supports the application's scalability and operational efficiency, making it adaptable to user needs and technological advancements.
Click to enlarge.
PROTOTYPEData collection enhances personalized bra fitting.
Users can choose from multiple data submission methods for their custom bra fitting, tailored to their comfort levels and abilities.
To promote a positive body image and encourage technology adoption, we provide inclusive filters during the interaction design process.
Prioritizing and conveying privacy in data handling.
Respecting user privacy is paramount. We ensure transparency from the start with a clear privacy policy screen at the beginning of the data collection process. Users are also encouraged to submit data with Wi-Fi turned off to educate them on on-device processing (edge computing), and to emphasize our commitment. And options for camera preferences and anonymous user accounts are also provided to enhance comfort and trust.
Visual analytics build trust and accuracy.
Our 'Analyze' page displays processed data through visualizations and charts, allowing users to see the accuracy of their bra fitting and recommendations. This page is designed to build trust and demonstrate our technology's potential without overwhelming users with excessive details.
Insights page educates and empowers.
The 'Insights' page allows users to track changes in their breasts and bra needs over time. It serves as an educational tool to help users understand, accept, and embrace their body changes. Features include customizable dashboards for tracking monthly or longer-term body responses.
Custom-fit bras: From virtual try-on to purchase.
After analysis, users can shop for and purchase recommended custom-fit bras. The virtual try-on feature lets users see how each bra would look and fit, ensuring confidence in their purchase.
Helping individuals find fit with their bodies, identities, communities and cultures.
Design carries political implications, necessitating accountability among designers for the norms, biases, and assumptions embedded in their creations. To offer bras that truly fit, it's crucial to empower users to define their own criteria for fit, challenging and destigmatizing prevailing notions of beauty, sexuality, and gender. This approach also serves as a valuable resource for individuals to deepen their understanding of their bodies, fostering body-positive mindsets and behaviors while enabling alignment between one's body, identity, and societal perception. This project has profoundly shaped the author's perspective on design and the role of designers, emphasizing the importance of inclusivity, well-being, and personalized intersections of mind, perception, and body, especially in addressing identity-related disparities in health outcomes.
ASIM Founder
Mar. 2022 - Present
ASIM aims to revolutionize the bra industry with bespoke, custom-fit designs, embodying a mission of body positivity, inclusivity, and acceptance.
As the driving force behind ASIM, I spearheaded research endeavors, business development, networking, and digital modeling.
See press on the project here:
https://www.fastcompany.com/91076873/students-world-changing-ideas-2024
https://id.iit.edu/projects/building-a-better-bra/
https://id.iit.edu/story/fitform-bras-is-a-world-changing-idea/
Next steps:
I will engage new contacts recently made whose work centers around the future of digital manufacturing, to add to the ASIM’s network, keep abreast of industry trends, and connect with like-minded individuals who share my passion for 3D knitting technology and generative design.
To fine-tune ASIM’s approach, I will conduct a second learning cycle survey to gain deeper insights into our service model’s desirability and customer value perceptions.
The computer vision and AI model is undergoing continuous development to improve accuracy, precision, and computational efficiency, I also plan to recruit participants for 3D scans, to better validate the AI’s output and develop a library of high quality 3D meshes for training the AI model.
In order to further advance research and development, I plan to continue with user and industry stakeholder interviews while also applying for grants, fellowships, and funding.
I will soon be contracting a fabricator in Michigan to knit the first high fidelity, 3D knitted prototype. This will additionally serve as an important test of the ASIM data to garment pipeline, and ensure that all output files are compatible with manufacturing process.
Appendix
Technology, Business and Service Components
The following highlights feature the technical aspects that the service model and business concept are built on, or considering.
Data capture and modeling for privacy.
Many collection methods were concept tested, but photogrammetry shows the most promise given its sensitive capabilities.
Photogrammetry, a computer vision method for measuring the physical world with photography, can be used to construct a 3D model of a body without invasive and shame-inducing direct measurement with calipers and measuring tape.
3D meshes are created based on how light and shadows fade into one another across skin and clothing.
Algorithms are designed for clothed and unclothed photographic inputs, and all processing is done on device to preserve privacy.
The camera experience is stylized visually, with data visualization overlays, to educate users about their bodies and minimize the risk for negative self-thoughts.
AI model identifies important anatomical “landmark” points in the 3D meshes, which are used for the next stage in computational analysis.
Analysis and simulation for accurate fit.
Computational physics simulation is a tool that is used to apply life-like physical forces to virtual objects.
The simulation stretches the skin, and moves breast tissue and internal glands against anchoring bones and muscles.
Designed as a collaborative simulation, users can add their preferences, learn about their breast anatomy through analytical visualizations, and compare alternative designs in real-time on a digital reconstruction of their body.
The simulated state reflecting the desired users preferences is what the bra generating algorithm aims to support.
Reddish colors identify where there is significant difference between location and scale of tissue.
Yellowish colors indicate where skin would best accommodate 3D repositioning.
Blueish colors show places where tissue needs direct support and should remain anchored in place.
This type of simulation can aid analysis in determining ideal position for breast support and symmetry in bra cup development.
Generative AI for bra actualization.
Generative AI algorithms integrates the individual needs and preferences with the physical and perceptual data and transforms it into a custom bra that supports those inputs.
The AI designs cater to various needs, and each option integrates precise anatomical measurements and style preferences, ensuring bras are both functional and reflective of individual identities, not only for the cup design, but the bra as a whole.
3D knitting digital fabrication for accuracy and efficiency.
3D knitting is a novel method of manufacturing complex 3D objects from specifiable threads.
Unique objects can be produced just as easily and reliably as mass manufactured items.
The knitting process is completely automated, constructing garments as one piece with no seams or other fabrication artifacts.
There is zero fabric waste, since only the thread needed to produce the garment is used.
Garments can be produced on-demand, reducing lead times and minimizing the need to warehouse products — ideal for entrepreneurial scaling and unpredictable demand cycles.
The process was highly researched, and a local manufacturing facility was visited and collaboration opportunities discussed. Upon receiving an order, the 3D digital models could be directly received and created by the machine.
