Set up for ITP Winter Show 2017:

LEVEL 1 – Follow the sin ! LEVEL 2 – Hold your breath!LEVEL 3 – Random!


Special thanks to our professors: Jeff Feddersen, Tom Igoe, and Daniel Schiffman for their advisement and feedback. Big hugs and kisses to Leon Eckert for helping us with all our coding inquiries, Aiden Nelson, and all our dear friends at ITP who tested our many, many prototypes along the way.

Wearable Enclosure

I thought last week’s class about rulers was my favorite until Ben taught us about enclosures. It’s always a wonderful treat to listen to someone talk about their passions with enthusiasm. It helps to put things into perspective and learn to appreciate them.


Leftover Acrylic  – Canal Plastic

Arcade Joystick – courtesy Roland

Bamboo Tray – Container Store

With finals in full swing, I planned on taking it easy and making the classic Luisa Pereira  enclosure to mount my joystick and house my Arduino. That didn’t come to fruition, however.

For PCOM/ICM finals, I am working with Simon Jensen on a wearable Heartbeat and Respiration monitor. On par with our timeline and this week’s user testing, it was time to enclose our respiration and heart sensors into a wearable. On Monday, I made a first prototype. Originally, I wanted to begin building it into a sports bra / tank because I want to begin thinking of the components as parts of a whole and understand how best to integrate them in an intuitive way. As a patternmaker, two of my primary concerns are fit and closures (how one gets into/out of a garment). I hit a roadblock thinking about how best to accommodate as many people as possible for user testing without having to make multiples. (Sizing standards exist for a reason and no size actually fits all.) Simon convinced me to focus on integrating the components first instead:

The design is essentially an adjustable belt that clips into position beneath the chest and above the abdomen. It has 2 channels for the respiration sensor and Polar belt to feed through:

It also has a pocket bag to house our feather board, LiPo battery, and bluetooth module. This is a video Simon documented of me wearing it beneath my shirt and testing it with data visualization of my breath:

I got great feedback from Simon and Aiden regarding size, comfort, sensor placement, and the adjustable strap closure. On my commute home, I sketched a design between falling asleep:

The next morning, I went to work and during my break, I drafted a pattern on CAD based on the measurements and feedback I received:

The pattern itself is quite simple. It is a rectangle on fold with 3/8″ seam allowances, clean finished edges and the channels are created by applying a topstitching. The top channel (1 1/8″) is for the respiration sensor, followed by an 1 1/2″ spacing between it and the Polar sensor channel to pocket the hardware. Notice the topstitching stops in the middle for alligator clip wires to attach to the respiration sensor.  I made a quick prototype out of muslin:

The trick to get all clean edges is by sewing an L-shape, flipping the remaining side through the pocket hole to sew together, then flipping it inside-out, then altogether right side out.

 Topstitching applied to create the channels:

I made some adjustments to the pattern and channel widths and remembered I had some leather leftover from a previous project:

Sadly, when I tried applying the final topstitching using the awful home-sewing machine in the ITP Soft Lab, it tore the leather. I know I’m not supposed to blame the machine and I know they aren’t equipped to sew leather to begin with, but I think our Soft Lab needs some upgrades. I almost lost my shit, but there’s no point moping and since I’ve already made three, I knew the fourth would come easier:

Back view w/ all parts enclosed:Front view w/ all parts enclosed:User testing today had a 6/7 success rate and received a lot of good feedback on our progress!

PCOMP/ICM Final: Project Planning + Respiration

Project Planning



Respiration Prototype Testing:

Last week, Simon and I met with Dan O’Sullivan and Simon’s PCOMP professor, Tom Igoe to get some feedback/advisement on our project idea. We were advised to rethink ways to incorporate more interactivity with our user and perhaps give focus to factors we can consciously control such as breathing, as opposed to heart rate, that we cannot unless accompanied by an universal trigger.

On Friday night, we tested the EeonTex conductive stretch fabric we ordered a week ago that we want to incorporate in our wearable to be used to measure respiration (and to be tested as a potential alternative to electrodes with adhesive backing). We tested its conductivity and stretch with a multimeter, then cut two 1/2″ x 13″ strips held together with safety pins to wrap around the chest. As we moved the alligator clips closer to the sternum at about 1″ ~ 1 1/4″ apart, we were able to visualize via serial monitor a consistent variation of ~30 points from deep breaths in and deep breaths out. We tested on ourselves, Sam, MH, and Leon.

To visualize our data in P5, we sat down with Leon to program the data to adjust accordingly to each individual.

My task this weekend was to make a working and adjustable prototype to playtest on Monday. I made and tested many to optimize data visualization. Different lengths, widths, layers, stitches, and combination of fabrics, but what we found in testing each prototype that followed was that our first safety pin prototype worked best.

The biggest challenge is working with stretch fabric with poor elastic retention. Every time the fabric is stretched out, it grows. That means we need to design a closure that not only accommodates people of different chest widths, but also accounts for fabric growth over time. This was a headache because I would test on myself and it would work perfectly, but if I tried to test on myself again after Simon, Jesse, and Kai, it would be too loose and noise would appear on screen. After many bouts of momentary successes followed by failure and frustration, I think I managed to make an adjustable prototype by repurposing some clasps from a backpack someone left on the junk shelf. The real challenge will come when we attempt to incorporate it in a garment, but for now, I have documentation it worked last night:

PCOMP/ICM Final Project Proposal: Wearable ECG

In my first year of working in the fashion industry, my supervisor told me, “It’s just clothes. You’re not saving lives.” This sentence continues to have a profound and resounding influence on me. On one hand, its repetition can provide a sense of relief when I’m confronted with challenging obstacles and demanding deadlines. In times of quiet introversion, it has the adverse effect of propelling me into existential crisis. Last year, I attended a panel discussion about brain-machine interfaces during the World Science Festival where neuroscientist Miguel Nicolelis gave a talk about his Walk Again project, a robotic exoskeleton aiming to restore full-body mobility to patients who suffer from paralysis. Further research opened my eyes to the potential of wearables. Learning my professional background could be useful in designing/redesigning assistive wearable technology, I applied to ITP to learn the technology.

For final project, I am working with Simon Jensen on a pulse and respiration sensing interactive wearable designed to make the ECG (electrocardiography) procedure more efficient and “seamingly wireless” with the incorporation of conductive soft materials (threads and fabric) and bluetooth. Instead of getting hooked up to a series of wires, we want the user (i.e. patients, athletes, etc.) to be able to wear a garment with the technology built in. We will use P5 serial control to visualize the data we collect from a person’s pulse using an ECG sensor and breathing patterns using a stretch sensor. The data will be synced to the beat of a  P5 sketch of an anatomically correct heart and diaphragm expanding and collapsing with every breath:


Last week, we successfully replaced ECG wires with conductive threads and metal snaps!

Over the weekend, we made our first prototype to understand whether incorporating our set up into a shirt might affect our data. We learned a lot, particularly with what power sources to use/not to use (coin cell batteries – although lightweight, are not reliable to power our multiple components – Arduino and ECG sensor). Although we got fairly accurate pulse readings, it only worked when the user was standing/sitting still. Movement distorted and added too much noise to the data. We will do more research on how to eliminate noise and filter accurate data. More to come…

Saturday, 11/04 – Our sweet classroom set up

We used one of Simon’s old shirts to make our 1st wearable prototype

Simon wearing prototype and resoldering some questionable ports

PCOMP Wonder Woman

When Jeff told us about a month ago that we would be paired up to do a Halloween themed midterm project, I knew it was the perfect opportunity for me to self-indulge in something I’ve wanted to make for a long time: a piece of clothing that lights up. In my costumes research phase, a lot of images that came up were of superheroes. I knew immediately I wanted to use Wonder Woman’s corset as a source of inspiration:I started prepping the dress form by taking measurements and taping the center front and back lines, bust, across back, and low hip lines. These lines give me visual cues when I drape to ensure everything is level and balanced. Using the reference photo above, I then taped down style lines from the corset onto the dress form and began to drape my muslin:


I did not want to be faithful to the original costume because often where fashion is combined with technology, it is easy for it to become theatrical and kitschy. That is never the direction/association I want for my work. I want to make beautiful garments that stand on their own.

After the half muslin was complete, I placed a string of lights to the front bodice for visualization. I then unpinned the muslin and moved onto transferring the muslin to paper patterns. A friend of mine who works in the bridal industry generously supplied several yards of beautiful silk organza. In preparation, the organza is pressed with a steam iron and laid flat on fold to cut a full bodice. The patterns are placed on grain and laid out strategically to optimize fabric, then cut out to be sewn:

I made the decision to expose the seams for a more visually interesting and 3-dimensional effect.


Next is arranging the strings of lights over the bodice (1 string per panel) by applying a number of handtacks.


With the corset complete, it was time to move to programming. When I found out I got paired up with Max, I was excited with the possibility of adding a musical component to our project. I looked into the Wonder Woman theme song and learned that the composition evokes power because it uses the same progression of notes and technique of dissonance to resolution as Robert Plant’s iconic opening wail in the Immigrant Song. More analysis in this video:

Wonder Woman Theme – Why It Evokes Intense Power

I pitched this to Max and we decided to sync each tone to a panel of lights, and have all the panels light up on the fifth note. I soldered extension wires to the cathode/anodes of the lights and we cobbled together some code to test the light and sounds:

int buttons[6];
//set up an array with 6 integers
//int buttons[0] = 2;
//give the first element of the array the value 2
int notes[] = {165, 196, 233, 247};
int led1 = 9;
int led2 = 10;
int led3 = 11;
int led4 = 12;
void setup() {
pinMode(led1, OUTPUT);
pinMode(led2, OUTPUT);
pinMode(led3, OUTPUT);
pinMode(led4, OUTPUT);
void loop() {
//  int tempo = analogRead(A1);
//  Serial.print(tempo);
//  Serial.print(“, “);
  int keyVal = analogRead(A0);
  if(keyVal >= 951){
    tone(8, notes[0], 200);
    digitalWrite(led1, HIGH);
    tone(8, notes[1], 200);
    digitalWrite(led1, LOW);
    digitalWrite(led2, HIGH);
    tone(8, notes[3], 200);
    digitalWrite(led2, LOW);
    digitalWrite(led3, HIGH);
    tone(8, notes[2], 2000);
    digitalWrite(led3, LOW);
    digitalWrite(led4, HIGH);
    digitalWrite(led4, LOW);
    tone(8, notes[3], 1000);
    digitalWrite(led1, HIGH);
    digitalWrite(led2, HIGH);
    digitalWrite(led3, HIGH);
    digitalWrite(led4, HIGH);
  else if(keyVal >= 51 && keyVal <= 950){
    tone(8, notes[0]*2, 200);
    digitalWrite(led1, HIGH);
    tone(8, notes[1]*2, 200);
    digitalWrite(led1, LOW);
    digitalWrite(led2, HIGH);
    tone(8, notes[3]*2, 200);
    digitalWrite(led2, LOW);
    digitalWrite(led3, HIGH);
    tone(8, notes[2]*2, 2000);
    digitalWrite(led3, LOW);
    digitalWrite(led4, HIGH);
    digitalWrite(led4, LOW);
    tone(8, notes[3]*2, 1000);
    digitalWrite(led1, HIGH);
    digitalWrite(led2, HIGH);
    digitalWrite(led3, HIGH);
    digitalWrite(led4, HIGH);
  else if(keyVal <= 50) {
    digitalWrite(led1, LOW);
    digitalWrite(led2, LOW);
    digitalWrite(led3, LOW);
    digitalWrite(led4, LOW);
We were able to get it to work with 10~15 minutes to spare before the Halloween Party. Today, Max took the code and ran with it to create a more interesting light & sound composition. He is a talented musician and it was a pleasure to collaborate on this project together. We look forward to working together again on future projects. This is our finished-for-now product:
(We will properly make video documentation tomorrow)
For this piece, I want the references to Wonder Woman to inspire and empower more women to pursue STEM fields.

Serial Input to P5.js

This weekend, I tried serial input to p5.js. Given my background in fabrication, programming is an abstract concept and language for me. For the past month, I had this nagging discomfort in the back of my mind every time I’ve had to open up p5. A few weeks ago, I finally figured out what it is: functions looping infinitely frame by frame gives me the same anxiety and impulse to shut a running faucet. This is why when Schiffman introduced serial to us in class last week, I actually breathed a sigh of relief. At least with my Arduino, I can physically unplug to stop all the madness. My homework’s not due until Wednesday, but I was excited to get started.

I watched the Serial Output from Arduino video from Tom and Jeff and used Schiffman’s simplified serial example code to get it working. At first, it didn’t seem to work and I kept getting “NaN” in my canvas. I noticed that in the same lab for Physical Computing, the code for including p5.serialport.js in the index is slightly longer than Schiffman’s simplified code and wondered if that was the problem.

After I replaced

<script src=”p5.serialport.js”></script>


<script language="javascript" type="text/javascript" src="p5.serialport.js"></script>

The sketch worked fine with each turn of my potentiometer and I was finally in control of the size or position of the ellipse on 
my screen. 

I later tried running the simplified code and it worked as well, so I'm not certain if there is any correlation or if there was just a lag 
when I tried it the first few times.

Week 3 Labs: Digital In/Output

Buttonswitch and potentiometer

Touch sensor

Servo – I got really frustrated because my servo wasn’t doing what my code was telling it to do. It jerked awkwardly and continued to do so even without my input (touch). I adjusted the code to no avail, and finally asked Krizia if I could test her’s. The code worked fine with her’s so I wasted close to two hours thanks to a defective servo.

Week 3: Observation


The technology I’m observing is the iPad Point-of-Sale (POS) stand at my favorite cafe in the city (and conveniently located a few blocks away from NYU): La Colombe. The set up is as follows:

  1. One person mans two check-out stands at the counter
  2. When one customer is signing for their transaction, the next customer in line is waved over to the other stand to process their drink order
  3. Meanwhile, baristas are taking orders from 2~3 people in line to maximize efficiency
  4. Two baristas to execute easier drink orders (drip coffees, ice coffees, etc.) and 1 barista on more complicated drinks (cappuccinos, lattes, espressos, etc.)

The technology perfectly complements the context of serving New Yorkers with wallets out in advance and need their caffeine fixes fast. The employees keep the lighthearted small-talk to a minimum and the transaction process is smooth and transparent. After punching in the order, the employee swerves the iPad around to face the customer to confirm the charge, tip amount and sign for the transaction. The average transaction time for customers paying with credit cards is approx. 17 seconds.

Fantasy Device: Empathy Patch

A NUI sticker/patch that you can attach to your head (small and easy to conceal in your hair) that is programmed to read your brain waves when it detects high “…activity in the nucleus accumbens and medial orbitofrontal cortex, two regions of the brain related to reward and higher executive function”. High activity in these regions are associated with empathic care. When this happens, you will receive a notification on your smart device, and encourage you to act in compassion. Naturally, when we act in compassion, we are happy, warm and tingly inside. Here is the link to the research. I coded a potential design for the patch. It’s inspired by schematic diagrams, which I thought would be fitting for the class.