Вы находитесь на странице: 1из 31



Issue 1
April 2019
Letter from the Editor | pg 3
Neusha Kharrati

World map of observe's reach | pg 4

Machine Learning: Modern Day

Artificial Intelligence  |  pg 5
Alex Zavalny

Should Designer Babies Be

Allowed?  |  pg 12
Helia Pashmforoush

Circuit Theory and Intuition for

Physical Equivalence  |  pg 17
William Zhang

Artificial Intelligence within

Medicine: Promising Diagnosis Success
Rate  |  pg 19
Isabella Polsfuss

3D Bio-printing: A Way Out for Organ

Shortage  |  pg 22
Kyle Hui

WDS 05599-3703B 1044AD | pg 26

Neusha KHarrati
Letter from the Editor
Dear readers,

It is with great pride that I launch Observe

International Science Magazine. I want to thank all of
the contributors who dedicated time to writing articles
and recruiting other students to spread the word. I have
been collecting articles since the beginning of this year
and will be publishing once a month to showcase the
work of the talented scientists and authors who took
part in this effort.
I first got the idea of starting a science magazine this summer at CSSSI, a
cardiothoracic surgical skills internship I participated in at Stanford. Throughout the
course of this internship, I met so many interesting and diverse students who
shared a passion for learning and science. I remember hearing interesting stories
exchanged about personal research experiences and concepts students found
interesting in class, and this was when I first considered creating a platform off of
which to share these stories.
Observe was founded to be a center of collaboration where likeminded
individuals could be exposed to the interests of others. As my idea began to
evolve, so did the extent of Observe’s network. I was curious to compare interests
not just among individuals at my internship but throughout the city in which I live,
my country, and eventually across the globe.
With the assistance of social media and word-of-mouth, Observe has expanded
around the world, with participants from the United States to the United Arab
Emirates to China. I am publishing Observe digitally so that individuals everywhere
can access it. I look forward to seeing Observe grow and hope that it makes
scientific discourse more accessible to high school students.

Neusha Kharrati
Editor of Observe Science Magazine
Observe's Global Reach
Machine Learning: Modern 5
Day Artificial Intelligence
Alex Zavalny
Malvern Preparatory School, Malvern Pennsylvania

The following paper discusses Machine
Learning in relation to being the most
influential part of A.I. (Artificial Intelligence),
and how it’s had an impact in the modern
day. Firstly, by providing historical context
of A.I., and how it lead to the creation and
application of Machine Learning. Secondly,
giving a brief technical explanation of
Machine Learning and how it’s used by
companies and businesses. Thirdly, how
companies like Google and Apple currently
use Machine Learning in order to improve
their line of products. With all of this, it can
be concluded that Machine Learning is,
without a doubt, an important part of our
lives and our future.

Modern Day Artificial Intelligence

Until nearly twenty-one years ago, it Deep Blue was still hardcoded to only
was deemed impossible to beat the play specific chess matches; it also
world renown chess grandmaster, Garry lacked the ability to adapt from its
Kasparov, when something finally mistakes unlike systems built on
defeated him. It was International Machine Learning (Demis, 2017).The
Business Machines’ supercomputer, following paper discusses the relevance
Deep Blue, which finally beat Kasparov of Machine Learning and how it’s the
in the nail-biting worldwide match in most important application of modern
1997. While still a huge achievement in Artificial Intelligence. First, by retelling a
the field of Artificial Intelligence (A.I.), brief history of Machine Learning from
the ancient Greeks, to the Turing Machine, Learning, is currently used to help
and leading up to the creation of Artificial companies and businesses. Before the
Intelligence and Machine Learning. Second, creation of Machine Learning or even A.I.,
a brief technical explanation of a lot of ancient stories, theory, and
Machine Learning, its philosophy first sparked the interest of
related fields, and how it relates to the many computer scientists to later join the
Advanced Placement Computer Science field in the mid 20th Century.
curriculum. And third, the multitude of ways
Machine Learning, specifically Deep

Historical Context
Machine Learning is a subset field of This universal calculus of reasoning is
Artificial Intelligence as defined by currently used by algorithms in Machine
Market Researcher and Social Media Learning systems in order to logically
Analyst Sanke Garbhe. Artificial calculate and make decisions (A.A.A.I.,
Intelligence’s earliest traces can be 2013). Then, in 1939, came the creation
found in the ancient Greek myths of of the famous Turing machine. Inspired by
Hephaestus, who would regularly build mathematician Kurt Gödel’s
himself mechanical servants to help him Incompleteness Theorem, that there are
manufacture weapons for the gods of certain statements believed to be true but
Mount Olympus (A.A.A.I., 2013). Also, in cannot be proven, Turing wanted to
Greek mythology was a giant made of create a machine that would be able to
bronze, named Talos, who was prove statements through the use of
considered to be an intelligent robot that algorithms (Smith, McGuire, Huang, &
guarded the island of Crete against Yang, 2006). The Turing Machine would
pirates (A.A.A.I., 2013). Fast forward into be able to do this using a reader that
the 17th century, philosopher René reads symbols on a long piece of tape,
Descartes proposed a theory, that and its sequence is used to build an
shocked all philosophers during this algorithm (Smith, McGuire, Huang, &
time, which was that all animals, Yang, 2006). These researchers from the
including humans, were just complex University of Washington also said that
machines and nothing more (A.A.A.I., Turing’s algorithms laid the foundations
2013). In that same century German for a framework of how algorithms in
mathematician, Gottfried Leibniz, had Machine Learning systems would work.
co-developed the Calculus Ratiocinator Demonstrating the power of the Turing
(A.A.A.I., 2013). It was a theoretical Machine, Turing was able to decipher the
universal calculation framework where previously undecipherable codes of
arguments could be solved mechanically Enigma by the Nazis.
(A.A.A.I., 2013).
Based on the practical uses of
and the human (Smith, McGuire, Huang,
algorithms, Turing later hypothesized an & Yang, 2006). Researcher Darrell West
idea that machines could think and solve from Brookings Institution said that the
problems in the same way humans promising results of the Turing Test had
would (Smith, McGuire, Huang, & Yang, given inspiration to computer scientist
2006). The researchers at the University John McCarthy. He then took the
of Washington also said that the creation development of the field a step further
of Artificial Intelligence began in 1950 by being the first scientist to coin the
when Turing published a paper, titled term Artificial Intelligence. West also said
Computing Machinery and Intelligence. that McCarthy described it as how a
The paper posed the question, can computer is capable of completing tasks
Machines think, and it described a in the same way as a human does. West
method to evaluate how a machine could also gives credit to computer scientist
think. This method was called the Turing Arthur Samuel for inventing the idea of
Test, and it proved that a machine is Machine Learning. His idea,
able to think by reasoning and revolutionary at the time, was to
completing logic puzzles similar to automate a computer to learn from data.
humans. The test itself was asking a This began Machine Learning which is
human and a computer a question, and currently an enormously important field
trying to determine which of the two of A.I. says Leading Instructor at
responses came from the computer Eduonix Learning Solutions, Sammy

Technical Explanation
Artificial Intelligence instructor Sam Supervised Learning is a predictive model
Eduonix defines Machine Learning as type where examples of input and output
programming algorithms to steadily get data are given in order to make a
closer to perform exactly how the prediction; this type of learning is what the
programmer wants them to perform. He Linear Regression example described.
says it also can be described in a similar Unsupervised Learning is a descriptive
process to Linear Regression; a random model type, where the type of input isn’t
line is drawn, and the error, the distance given, and associations must be made
from each point to the line, is recorded. for the types of output data. An example
Then, new lines are continuously drawn Eduonix gave was a supermarket
until they reach a point where the error is associating the types of products to the
minimal and the line best describes the types of people buying them.
points. Machine Learning is made up of 3 Reinforcement Learning has similar
parts which are Supervised, techniques of association in Unsupervised
Unsupervised, and Reinforcement Learning, but it instead performs in a way
Learning (Eduonix, 2018). Statistically, to maximize a reward
(Eduonix, 2018). given, and associations for more association to be done from
must be made for the types of output node to node, and at the same time
data. An example Eduonix gave was a having more nodes allows the algorithm
supermarket associating the types of to do more advanced learning. This
products to the types of people buying allows Deep Learning to train from
them. Reinforcement Learning has extremely large amounts of data, proving
similar techniques of association in itself to be a very useful tool for
Unsupervised Learning, but it instead companies said Sammy Eduonix.
performs in a way to maximize a reward According to the College Board, the
(Eduonix, 2018). Machine Learning also Advanced Placement Computer Science
incorporates the use of Neural Networks curriculum includes many topics of
which are structures based on the programming such as data structures
neurons and synapses in the human and different types of sorting and
brain (Eduonix, 2018). They are searching algorithms. In fact, these
structured in layers comprised of nodes topics are the fundamental ideas that the
that hold data. Then, data from one node fields of A.I. and Machine Learning are
is sent to another node in the next layer based on. The structures of Neural
in a movement similar to the motion of Networks holding nodes of variables are
how an algorithm in a Turing Machine similar to arrays and array lists holding
works. The first layer is the Input Layer, primitive types and objects. Some
the last layer is the Output Layer, and the algorithmic concepts learned in A.P.
layers in between are the Hidden Layers. Computer Science include conditional
Sam Eduonix describes them as the statements, looping, and combining the
most crucial part to a Machine Learning two to make algorithms that analyze
algorithm because it’s where most of the data. Algorithms that parse through
association from node to node is done. In items in an array are an example of how
normal Neural Networks, there is usually data moves through the Hidden Layers
one Hidden Layer, however, the famous in a Neural Network. These algorithms
concept of Deep Learning explores the are the key for large companies to be
use of multiple Hidden Layers to be able able to learn and predict from large
to associate with many, many nodes. amounts of data in order to improve
Having multiple Hidden Layers allows themselves according to A.I. instructor
Sam Eudonix.

Future Relevance
In the modern day, the amount of data internet, databases, and servers Bernard
humans generate daily is estimated to be also describes how important it is for
2.6 quintillion bytes says Business and companies to be able to parse through
Technology Advisor at Forbes, Marr that enormous amount of data. He also
Bernard. Through the further mentions how major companies such as
development and refinement of the Google and Apple use Deep
Learning, specifically Supervised OnePlus 6, LG’s G7, and Motorola’s 9
Learning and Unsupervised Learning, in Moto G6 are all examples of flagship
order to gather patterns and make phones using facial recognition as a form
inferences about their enormous amount of identification in order to unlock the
of data. This allows them to analyze, device. That means that most people
interpret, and make predictions on their carry many Deep Learning programs in
data. Researchers from Harvard their pockets every day without even
Business Review said that companies realizing it. Along with companies taking
like the accounting network organization, advantage of Machine Learning with
Deloitte, use Deep Learning to “extract speech and facial recognition, other
terms from contracts, which enables an products and tools used by people every
audit to address a much higher day take advantage of it as well. Some of
proportion of documents, often 100%, the most used areas in Machine
without human auditors having to Learning include handwriting recognition,
painstakingly read through stock market analysis, news
them”(Davenport and Ronanki, 2018). categorization, search engines like
The researchers also said that other Google, and weather prediction (De
applications of Machine Learning by Spiegeleire, S., Maas, M., & Sweijs, T.,
businesses include predicting what a 2017). In 2016, Google switched to using
customer is likely to buy, identify credit a Deep Learning algorithm with Google
and insurance fraud in real time, Translate. This switch caused a 60%
identifying safety or quality problems in decrease in errors, according to the
manufactured products, and automating Google computer scientists who worked
the process of targeted advertisements with the new algorithms (Castelvecchi,
to the user. In addition to data analysis, Davide.). With all this, it’s pretty evident
another major use of Machine Learning that Machine Learning is extremely
is with modern virtual voice assistants important, because of its enormous
which can analyze, recognize, and impact.
understand human speech (Bernard, For over seven years, Game of Thrones
2018). Bernard also found that Google’s fans have been waiting patiently for
Google Home, Apple’s Siri, Amazon’s author George R.R. Martin’s next book in
Alexa, and Microsoft’s Cortana are all series: The Winds of Winter Since.
virtual voice assistants, used daily in Impatient but eager software engineer
most modern homes, that are Deep Zack Thoutt is currently training a neural
Learning based. Many of the modern network that predicts the events of the
phone companies use facial recognition next book, which is an example of
on their newest cell phones, and they’re Supervised Learning (Hill, 2018).
also built on Machine Learning Although the network isn’t sophisticated
algorithms (Bernard, 2018). Apple’s enough to write a complex narrative, it’s
iPhone X, Samsung’s S9 and S9 Plus, a step into a world where A.I. is able to
Samsung’s Galaxy Note 9, Window’s
write as well, if not better, than humans businesses.They also continue to 10
(Hill, 2018). And who’s to say the very use Deep Learning in order to grow and
paper being read right now isn’t written to progress which allows for the creation
by a neural network? We can never of new technology to be made. With all
know for certain. Machine Learning isn’t the data and information present, it’s
a new concept and has been around clear that Machine Learning has given
since the mid 20th century with ties all and will continue to be a major impact on
the way back to the Ancient Greek myths humanity. Hopefully, the field of A.I.
of Homer. Machine Learning is made up progresses toward a future where
of Supervised, Unsupervised, and humans can become chess
Reinforcement Learning which are used grandmasters of the world like Kasparov
by major companies and before Deep Blue.

AAAI(2013)N/A(N/A). A Brief History of AI. Association for the Advancement of Artificial
Intelligence. i2k Connect. Retrieved from
Brogaard, Jonathan and Zareei, Abalfazl. (August 16, 2018). Machine Learning and the
Stock Market. Retrieved from
https://ssrn.com/abstract=3233119 or http://dx.doi.org/10.2139/ssrn.3233119
Browne, Ryan. (August 1st, 2018). Five of the Scariest Predictions for A.I. CNBC Tech.
Retrieved from::
Castelvecchi, Davide. (September 27th, 2016). Deep Learning boosts Google Translate
tool. Nature International Journal of Science. Retrieved from
Davenport, Thomas and Ronanki, Rajeev. (January, 2018). Artificial Intelligence for the
Real World. Harvard Business Review. Retrieved from
De Spiegeleire, S., Maas, M., & Sweijs, T. (2017). ARTIFICIAL INTELLIGENCE AND
MEDIUM-SIZED FORCE PROVIDERS (pp. 43-59, Rep.). Hague Centre for Strategic
Studies. Retrieved from http://www.jstor.org/stable/resrep12564.8
Eduonix, Samy. (Udemy). (November, 2018). Projects in Machine Learning: Beginner To
Professional. [Webinar]. Retrieved from https://www.udemy.com/machine-learning-for-
Garbhe, Anket.(November 3rd, 2017). What is Artificial Intelligence. Medium.com.
Retrieved from
Hassabis, Demis. (April 27th, 2017). Artificial Intelligence: Chess match of the 11
Nature International Journal of Science. Retrieved from
Hill, Sam. (August 28th, 2017). A Neural Network Wrote the Next 'Game of Thrones'
Book Because George R.R. Martin Hasn't. Motherboard.Vice. Retrieved from:
Lou, Harbing. (August 28th, 2017). AI in Video Games: Toward a More Intelligent
Game. Harvard’s Science in the news. Retrieved from
Marr, Bernard.(October 1rst, 2018).What Is Deep Learning AI? A Simple Guide With 8
Practical Examples. Forbes. Retrieved from
[Serrano, Luis]. (September 9th, 2016). A Friendly Introduction to Machine Learning.
[Video]. Retrieved from
Smith, McGuire, Huang, Yang. (December, 2006). History of Artificial Intelligence.
History of Computing, University of Washington. Retrieved from
Strauss, Karsten (Febuary 2, 2018). Role of Artificial Intelligence in the future of job
search. Retrieved from https://www.forbes.com/sites/karstenstrauss/2018/02/02/the-
West, Darrell. ( October 4th, 2018). What is artificial Intelligence? Brookings. Artificial
Intelligence and Emerging Technologies Initiative. Retrieved from
Should Designer Babies
Be Allowed?

Helia Pashmforoush

Torrey Pines High School, San Diego California

A designer baby
is a baby genetic
-ally engineered for
specially selected traits,
which varies from a lowered
risk of disease to gender selection.
Before the use of genetic
engineering, designer babies were primarily
a concept in science fiction. However, the
rapid advancement of technology before and
after the twenty-first century makes designer babies
an increasingly real possibility as already created by a
scientist in China. This new advancement in technology could
be used to change the looks of a child or prevent diseases. To
alter the appearance of a child is not bioethical, whereas altering a child's
genes to prevent harmful diseases is arguably justifiable.
CRISPR is a new technological and benefits of editing embryos. She
advancement used to alter the genetic supports this claim by first stating that
makeup of future generations and create CRISPR has made the engineering of
designer babies that are genetically embryos cheaper, then goes to explain
modified before birth as an embryo for how in most places it is not regulated to
appearances or genetic disorders. In the changes genes, then she mentions if
article Editing Embryos: Considering gene editing were to be used the United
Restriction on Genetically Engineering States would need to put up regulations,
Humans, 2015-2016, Anna Zeret and finally she claims that the use of
questions the safety gene modification without proper
regulation could cause danger to the reasonable solution to a long-lasting
individual and society. This technology problem passed down generations. In
has already been used in China by Genetic Inequality: Human Genetic
scientist He Jiankui when he leads his Engineering, Danielle Simmons argues
team using the gene-editing technology that “As genetics allows us to turn the
CRISPR to prevent disease in newborn tide on human disease, it's also granting
twins. the power to engineer desirable traits
A Chinese scientist created the first into humans (Danielle Simmons).” She
ever genetically altered twins using supports this claim by first, stating the
technology that is banned in the United reasons why someone would want to
States. His motive to create these genetically engineer a human, then
modified twins was to prevent any harm mentions how genetic modification had
that would come from having parents already been tried on animals, then she
with HIV. In the text, Chinese mentions that wealthier people may
Researcher Claims First Gene-Edited consider changing traits to make their
Babies, November 26, 2018, Marilynn child superior, and finally claims that if
Marchione claims that the use of the option to genetically modify children
CRISPR provides great potential for was possible, regulations would need to
correcting mutations and diseases, but be enforced. These new technological
could also be passed down generations. advances are beneficial for correcting
She supports this claim by first genetic disorders but could be harmful
explaining the purpose of CRISPR which by opening the door to genetic haves
allowed for the Chinese scientist to and have-nots without chances for a fair
create the genetically modified babies, and just future.
then going into the possibility that the The use of genetic modification has
genetically modified babies could pass its benefits of preventing diseases but
down their traits to future generations, may be used to change appearances,
then the scientist explaining as to why which is not bioethical. In the statement
he decided to edit the embryos which Claim of Genetically Modified Babies,
was to get rid of any possibility of a the Center for Genetics and Society
disorder from HIV, and finally how it claims that “throwing open the door to a
could be very helpful to get rid of other society of genetic haves and have-nots
genetic disorders. As this technology undermines our chances for a fair and
becomes more widely used, precaution just future (Center for Genetics and
must be provided to ensure that CRISPR Society).” The author supports this claim
is only used for genetic mutations rather by first explaining to us that “Jiankui He
than desired traits. With altering the has announced the birth of twin girls
appearance of a child comes bio-ethical whose DNA he claims to have altered
concerns as opposed to the alterations using the gene-editing technique
of genes to prevent harmful diseases, CRISPR”, then says that the benefit for
which may be a gene-editing would be the prevention of
diseases, then says how the experiment a pair of molecular scissors, capable of
that Jiankui He performed may have cutting strands of DNA which could
been unregulated, and finally mentions correct or alter the genetic code. In the
“It’s hard to imagine a graver abuse of a text, Genetically Modified Babies
child...If this goes unchallenged, other February 23, 2014, Marcy Darnovsky
rogue actors will soon offer wealthy claims that “We should carefully and
parents purported genetic thoughtfully apply the tools of human
enhancements for their children.” There genetic engineering to treat medical
are different types of gene alterations conditions in people, but we should not
that come at different high prices. use them to manipulate the genetic traits
Wealthier families may use CRISPR to of future children (Marcy Darnovsky).”
enhance their child's features. She supports this claim by first saying
The uses for CRISPR are that genetically engineered children
consequential if used for the wrong would pass down their new genes to
reasons such as improvement in skills. future generations, then going into how
In the article, Human Genetic many newborns have mitochondrial
Modification, the author claims that gene diseases that could have been
editing is “the most consequential type of prevented if using CRISPR, then
genetic modification. If used for explains how parents could prevent any
enhancement purposes, it could open harm coming to their child, and finally
the door to a new market-based form of claiming that the new technology should
eugenics.” The author supports this only be used to prevent any harm that
claim by first stating the two different may come to the baby. The use of new
types of gene modifications known as technology raises moral concerns about
somatic and germline, then explains how the safety and destruction of the embryo.
the costs of these treatments would be An issue that comes with genetically
very high, then says how if a modified modifying embryos is safety and access.
baby were to reproduce in the future the In Balancing Ethical Pros and Cons of
new modified traits would be passed Stem Cell-Derived Gametes January 13,
down, and finally mentions that this new 2017, Seppe Segers argues that “The
technology is outlawed “in more than 40 technology of SCD-gametes raises
countries, and by a binding international moral concerns of how reproductive
treaty of the Council of Europe.” This autonomy relates to issues of embryo
technology is very new, and not many destruction, safety, access, and
laws and regulations are put into action applications beyond clinical infertility
yet. If CRISPR becomes more widely (Seppe Segers).” He supports this claim
used, safety precautions would need to by first providing an overview of the
be established. ethical pros and cons of stem cell
CRISPR is an enzyme that acts like derived gametes, then stating that stem
cells hold promising applications
Designer babies aren’t futuristic. They’re
research and for clinical use in assisted already here. October 22, 2018, Laura
reproduction, then explores the “ethical Hercher suggests that designing
issues connected to gametes derived inequality in genes may occur. She
from embryonic stem cells (both patient supports this claim by first telling the
specific and non-patient specific)” and story of Matthew and Olivia and how
finally discussing the moral concerns they had to alter their embryo, then goes
that come with altering an embryo. into the description of the cost of these
Another concern with the modifying of procedures, then talks about how
embryos is the rights the child is born “Americans approve of using
with. reproductive genetic tests to prevent
When a Chinese scientist modified fatal childhood disease, but do not
the embryos of twins, many moral approve of using the same tests to
concerns came with how the newborn identify or select for traits like intelligence
child should be treated. In A Chinese or strength” and finally how the choice to
Scientist Says He Edited Babies' Genes. get rid of a disease is ideal to most
What Are the Rights of the Genetically people. The use of CRISPR to aid a
Modified Child? December 6, 2018, healthy pregnancy for the child has
Eileen Hunt Botting, questions that recently been an available option for not
genetically modified children may have many families, but in the future could be
different rights. She supports this claim widely accessible and used for the births
by first explaining the process of of healthier children.
modifying an embryo, then explaining The increased ability to control and
the reasoning behind why someone manipulate embryos broadens many
would want to alter a gene, then using possibilities for improving the health of
examples of movies such as “Gattaca” children, yet these possibilities are faced
that display modifying babies as a with potential social repercussions that
normality, and finally stating that children could have negative consequences in
with genome modifications have the right the future. Designer babies represent
to share love with parental guardians great potential in the field of medicine
and have no discrimination on the basis and scientific research and allow the
of birth. There are other cases of ability to treat children through prenatal
embryos being modified already in order diagnosis.
to keep the baby out of any danger.
Since this new technology has been
created, there have been other few
cases in which parents have been
allowed the chance to alter their
embryos to maintain the safety of a child.
Example from Source: In
Botting, Eileen Hunt. “A Chinese Scientist Says He Edited Babies' Genes. What Are the
Rights of the Genetically Modified Child?” The Washington Post, WP Company, 6 Dec.
2018, www.washingtonpost.com/news/monkey-cage/wp/2018/12/06/a-chinese-scientist-
Center for Genetics and Society. “Claim of Genetically Modified Babies: If True, a Grave
Abuse of Human Rights.” Claim of Genetically Modified Babies: If True, a Grave Abuse
of Human Rights | Center for Genetics and Society, 2018,
Darnovsky, Marcy. Genetically Modified Babies. 23 Feb. 2014.
Darnovsky, Marcy. “Genetically Engineered Children?” Genetically Engineered
Children? | Center for Genetics and Society, 2015,
“Human Genetic Modification.” Human Genetic Modification | Center for Genetics and
Society, www.geneticsandsociety.org/topics/human-genetic-modification.
Ly, Sarah. “The Embryo Project Encyclopedia.” Temperature-Dependent Sex
Determination in Reptiles | The Embryo Project Encyclopedia, Arizona State University.
School of Life Sciences. Center for Biology and Society. Embryo Project Encyclopedia.,
31 Mar. 2011, embryo.asu.edu/pages/ethics-designer-babies.
Marchione, Marilynn. “Chinese Researcher Claims First Gene-Edited Babies.” AP News,
Associated Press, 26 Nov. 2018,
Roger Lock & Colin Miles. “Biotechnology and Genetic Engineering: Students'
Knowledge and Attitudes.” Taylor & Francis, 2010,
Rosenberg, Dr.Eugene. “Genetic Engineering.” Genetic Engineering - an Overview |
ScienceDirect Topics, 2017, www.sciencedirect.com/topics/neuroscience/genetic-
Segers, Seppe, et al. “Balancing Ethical Pros and Cons of Stem Cell Derived Gametes.”
SpringerLink, Springer, 13 Jan. 2017, link.springer.com/article/10.1007/s10439-017-
Simmons, Danielle. “Genetic Inequality: Human Genetic Engineering.” Nature News,
Nature Publishing Group, 2008, www.nature.com/scitable/topicpage/genetic-inequality-
Zaret, Anna. “Editing Embryos: Considering Restriction on Genetically Engineering
Humans .” HeinOnline, 2015, heinonline.org/HOL/LandingPage?
Circuit Theory and
Intuition for Physical
William Zhang
Ruben S. Ayala High School, Chino Hills California
One of the most basic ideas taught in their study of circuitry is this 18
early on in the study of basic circuit theory idea of equivalence, which is not limited to
is Equivalent Resistance, which allows resistor values, but also to the structure of
one to decompose a complicated circuit the circuit itself. It is a mysterious idea at
into its simplest form. For those more first, that a seemingly impossible to
familiar with chemistry, I would liken it to analyze circuit can be manipulated into an
molecular formulas and empirical arrangement that, structurally seems very
formulas, where the empirical formula is a different, but physically is absolutely
“simplified” version of the molecular equivalent. A great example of this that I
formula, but that’s where the analogy learned of recently is the Delta to Y / Y to
ends. The idea of Equivalence in Circuit Delta transformation (figure A). Within the
Theory is much more useful to the context of the system, the conversion
engineer than empirical formulas are to doesn’t change the behavior of the circuit.
the chemist. This logic is also what leads to the
A bit of background to make sure we’re concept of Black Box circuits. Black Box
on the same page: The resistor is likely basically means that a circuit can be
the most ubiquitous component in circuitry. replaced by another circuit as long as the
In the real world, each resistor is assigned overall behavior is maintained. Imagine a
a certain resistance value that is basically circuit with two resistors A and B, with an
a measurement of how well it impedes the infinite amount of three resistors in series
flow of current. Ohm’s law relates the made parallel to B, as shown in (Figure
resistance value of a resistor with the B). Due to the way resistors in parallel
voltage across it and the current that are calculated, the equivalent resistance
passes through it. of the section parallel to Resistor B
Essentially, resistors placed in series or eventually approaches a certain value R,
in parallel to each other can be which finally allows the circuit to be
mathematically manipulated to create an analyzed.
Equivalent Resistor. Multiple resistors in Thankfully, calculus provides us with
series is equivalent to a single resistor with the mathematical foundation behind all
a resistance value that is the sum of the the murkiness. Calculus is what permits
resistance values of every resistor that these equivalent systems, as it sets up
contributes to it. The reciprocal of the the relationships between currents,
equivalent resistance of resistors in resistance, and voltage. While Ohm’s law
parallel is equal to the sum of the is a great tool for high schoolers, it
reciprocals of each resistance value in becomes useless when the values of
parallel. This enables one to convert a interest have rates of change, as in the
complicated looking circuit into a much real world. Therefore, differential
simpler form, turning some extremely equations are required. Equivalent
complicated systems into simple math and systems yield thesame differential
logic problems. equations. Thus, they produce the same
A common theme that one will notice solutions and exhibit identical behavior.
Artificial Intelligence within
Medicine: Promising Diagnosis
Success Rate
Isabella Polsfuss
Francis Parker School, San Diego California

in many of our minds as we approach

university and beyond.
Of course, though the answer to such
a question is in many respects
theoretical, multiple developments are
currently being made to transform what
was once speculation into reality. I
currently hold an interest in medicine as I
have realized pursuing a career in
healthcare provides me with the
harmonious opportunity to synthesize my
love of science with human interaction.
However, I have recently come across
many instances in which the success
rate of artificial
intelligence is higher than that of
Our understanding of the world is
professional doctors, whom have
becoming increasingly complex through
devoted their lifetime developing
the immense knowledge gained by new
knowledge in such a field.
technologies. As a member of Francis
One prominent example of rising
Parker’s robotics team myself, I am
artificial intelligence is within an
aware that the capabilities of our
operating system created by researchers
workforce is dramatically changing with
from China’s leading hospitals to self-
such innovation. Could a machine or
diagnose patients. In mid-2018, a
computer system do the job I wish to
competition was held between this
pursue? This is a question that our
system - known to the
generation is left to ponder, and it lurks
world as “BioMind” - against well-reputed These results are extremely 20
doctors. The results were astounding; exciting, and should not be discouraging
BioMind diagnosed 87% of brain tumors for those, like myself, who have an
correctly in approximately 15 minutes. In interest in perhaps pursuing medicine.
contrast, a team of 15 doctors achieved Dr. Wang Chongqing is one of many
66% success in 30 minutes (Dixon). This Polsfuss 2
artificial intelligence system runs under whom feels that “humans and machines
repetition; essentially, the developers are good at different things” and can
provided BioMind with tens of thousands “complement each other” (Tmg). Though
of past images covering nervous systems like BioMind may be more
system-related diseases. Such accurate than humans on some levels,
exposure-based algorithms have not there is an obvious lack of interaction;
only been applied to the complexities of they are unable to diagnose fully and
the nervous system, but also have been take into account important factors of the
utilized to make other diagnoses patient such as age, gender, and history.
throughout the human body. For The best, most simplistic comparison to
example, the presence or absence of this is like using a GPS to drive a car.
tuberculosis, diabetes, or even Though one may drive a car without this
possibilities of future heart attacks have system, it certainly is beneficial and often
all been studied using such methods prevents one from getting lost. The car
(Hsieh). cannot drive solely with the GPS, of
Another AI system researched in late course, so it will be the driver (doctor)
December 2018 has been found to whom ultimately makes the decisions
predict Alzheimer's disease (AD) years (Tmg). Thus, the rise of artificial
before diagnosis, which is similarly as intelligence within medicine shouldn’t be
promising. Diagnosis of AD is extremely feared, but rather accepted and
time-sensitive, yet also extremely promoted. Developing systems will likely
challenging. Glucose uptakes in certain not completely displace doctors within
regions of the brain are able to indicate a our generation; instead, it should allow
change in metabolism and show flags of leaders in healthcare to perform tasks
the disease, though are very subtle and more efficiently and expand to new
can easily diffuse. A research team research.
composed of physicians and engineers
utilized deep learning software and
trained the algorithm with over 2,000
brain scans. The performance of the
algorithm has a sensitivity enabling the
disease to be detected an average of six
years prior to the final diagnosis (“AI”).
“AI Predicts Alzheimer's Years Before Diagnosis.” RSNA,
Dixon, Robyn. “China Plans to Dominate AI, with a Vanguard of Robotic Doctors like
'Biomind'.” Los Angeles Times, Los Angeles Times, 6 July 2018,
Hsieh, Paul. “AI In Medicine: Rise Of The Machines.” Forbes, Forbes Magazine, 30 Apr.
Tmg. “Why AI Could Be the Very Best Second Opinion in Medicine.” The Telegraph,
Telegraph Media Group, 24 July 2018,
3D Bio-printing: A Way
Out for Organ Shortage
Kyle Hui
St. Paul’s Co-educational College, Hong Kong
As science advances worldwide,
patient-specific constructs. The 3D digital
technologies applied in healthcare model is then sliced into a number of 2D
sectors improve as well. This progress is horizontal layers, facilitating the additive
a double-edged sword. On one hand, printing process later on. Finally, the
stronger vaccinations and medicine gift bioink is printed onto scaffolds (may or
society with better public health. On the may not contain cells by the technique of
other hand, the decrease in availability of cell seeding) or scaffold-free molds.
organs can be devastating to those in Theoretically, an organ would then be
need of transplants. Growing up in the printed out. However, there are many
bustling city Hong Kong, a place difficulties and limitations during each
populated with over 7 million people, I step. For example, cells will die quickly if
have always been worried about how the not exposed to suitable environments in
healthcare system sustains the city. In the ink. In addition, many existing 3D
fact, Hong Kong is among the world’s printers are not powerful enough to
worst for organ donations, with only 5.8 create structures strong enough to
donors for every million people in the city. support the tissue.
Many patients in the city die because of Despite the numerous challenges
the absence of suitable organs. faced by scientists, researchers have
Having a strong passion for biology, I been taking small but exciting steps
keep my eyes open to solutions that can towards a future of artificial organs. In
alleviate this ever-growing problem in my late 2018, scientists from the University
hometown. Last summer, I took a course of Utah published a paper in the Journal
in Stanford on Tissue Engineering and of Tissue Engineering. In the paper, they
3D bio-printing. During the course, I had described their success in manufacturing
a sudden moment of eureka. Can’t we human tissues like ligaments and
just print more organs to put an end to tendons through bio-printing. The team
this crisis? Little did I know that the harvested stem cells from human fat and
answer to this question is theoretical in printed them onto a layer of hydrogel.
many ways. In fact, we are still quite a Over time, the structure developed into
long way from printing body parts like tissues resembling ligaments and
how we print paper. tendons. Since connective tissue is a
In essence, 3D bio-printing can be complex of different kinds of cells, these
broken into a few key steps. First, researchers developed a special
scientists concoct bioink. In order to printhead that could lay down specific
create this, they differentiate stem cells cells at specific locations in a controlled
into specific cells, then encapsulate them manner. This discovery was important
in hydrogels. Afterwards, medical because current tissue replacement
imaging technology is employed to methods rely on tissues from other parts
understand the 3D structure of organs, of the patient’s body. These tissues may
allowing medical professionals to design
not be of high quality and the harvested midst of all excitement, I believe that
locations may be damaged. The careful and small steps must be taken.
University’s research can effectively There are many ethical considerations
these issues and even serve to recreate on the technology of 3D bio-printing. Is it
spinal discs for transplants. morally right to create a new body part?
Another recent and ground-breaking Is it morally correct to harvest cells from
discovery is the printing of a half- an embryo to reprogram them?
functioning organ. In early 2019, Recently, Chinese scientist He Jiankui
researchers at the Tel Aviv University of claimed to have use CRISPR to disable
Israel managed to engineer a cherry- the gene for making CCR5 in babies,
sized heart using human cells. Like the making them immune to HIV. This has
scientists in the University of Utah, the caused uproar amongst the scientific
Israeli researchers took fatty tissue from community, saying that the study
humans and reprogrammed the cells to infringed on the babies’ rights, and that
become cardiac cells and blood vessel the technology was still unsteady at the
cells that know how to circulate blood time. Hopefully, scientists and the
through the cardiac system. Trapping government can work hand in hand,
these cells in bioink, they then printed taking slow but steady steps so that we
out the organ. What makes this can print organs safely and ethically in
discovery stand out is that the the future.
researchers succeeded in creating the
working blood cells to keep the organ
alive, something no other scientist could
before. Although the heart can contract,
the cells do not work in perfect harmony
to form a pumping ability. Moreover, the
current 3D-printers lack precision and
resolution that can accurately print the
tiniest of human blood vessels. This pint-
sized heart may still not yet be functional
in humans, but it marks a huge
milestone in biomedical sciences.
These results are amazing, pointing
us towards an era where we can literally
print out new ears or noses for
ourselves. The common medical
obstacle of immuno-rejection of cells can
also be eliminated in the future with
patient-derived organs. However, in the
“3D Printing Human Ligaments, Tendons and More . . .” State of Utah Center of
Excellence for Biomedical Microfluidics, 17 Oct. 2018, mems.utah.edu/2018/10/17/3d-
Cheung, Elizabeth. “Hong Kong among World's Worst for Organ Donations.” South China
Morning Post, 20 July 2018, www.scmp.com/news/hong-kong/health-
Cui, Haitao, et al. “3D Bioprinting for Organ Regeneration.” Advanced Healthcare
Materials, U.S. National Library of Medicine, Jan. 2017,
Lazarus, David. “Israeli Scientists Unveil World's First 3D-Printed Heart with Human
Tissue.” The Times of Israel, Apr. 2019, www.timesofisrael.com/israeli-scientists-unveil-
Park, Alice. “Chinese University 'Shocked' By News of First CRISPR Babies.” Time, Time,
26 Nov. 2018, time.com/5463741/crispr-human-babies/.
Astrometric Measurements
and Historical Data Summary
of WDS 05599-3703B 1044AD

Neusha Kharrati
Francis Parker School, San Diego California

Position angle and angular separation for the
double star system WDS 05599-3703B 1044AD
were measured from images acquired through
the iTelescope network, and processed through
Maxim DL and MiraPro x64. These astrometric
measurements were calculated to be 189.9º and
27.3". The results of these measurements,
compared to historical measurements

The selection criteria for determining a double
star system to study included a right ascension
between 00 and 06 hours, a separation of at least 6
arcseconds, and a maximum differential magnitude
of 6. The Washington Double Star
(WDS) catalog, managed by the United States Naval Observatory (USNO), was used to
find a candidate star system. Double star 05599-3703 B 1044AD (referred to as 1044AD
hereinafter) fit this criteria, with a right ascension of 05h 59m 55.79s, a separation angle
(Rho) of 21.3", and a differential magnitude of 4.58.
Historical data, provided by the USNO, shows two complete sets of previous
measurements. There are other measurements that did not provide position angle (theta)
or angular separation (rho), but instead Double Star 1044AD was
provided data such as magnitude for a selected with the objective of taking new
given epoch. The earliest measurement astrometric measurements adding to the
was made in 1999 indicating a theta of historical measurements in an effort to
215 º and a rho of 16.2". The other assist in determination of the system’s
measurement, also taken in 1999, shows gravitational nature, while also
a theta of 216° and a rho of 15.8". evaluating the possibility of a chance
alignment of the A star over the D star.

Materials and Methods

Images were taken through the magnitude between the A and D stars.
iTelescope network where each Because of the high magnitude of both
observatory has multiple telescopes that the A and D stars—which are G type
could be used for scientific research. stars, indicating that they peak in yellow
The first telescope used was the T27, —we wanted to look at them in a color
Figure 1, located in Siding Spring, they didn’t peak in so to see less light,
Australia at an elevation of 1122 meters. and consequently ended up using the
T27 has a resolution of 0.53 arc- blue filter. These different filters were
secs/pixel, a pixel size of 12um square, used to discern whether or not there
and a position angle of 0º. The second were differences in measurement that
telescope was the T30, Figure 2, also could be attributed to the varying
located at 1122 meters in Siding Spring, wavelengths corresponding to each
Australia. This telescope has a filter.
resolution of 0.81 arc-secs/pixel, a pixel Each image was processed through
size of 9um square, and a position angle Maxim DL, software for astronomical
of 086º. imaging which processes data acquired
We requested images, with select through imaging array detectors such as
filters, exposure length, and time CCDs (charge-coupled device). Mira Pro
windows for the images. The equipment was used to clean, label, and calibrate
used to measure our stars included 2 images, along with measuring our
images of Luminance with 30 seconds position angle and separation. To ensure
exposure, one Luminance with 60 accurate measurements in Mira Pro, the
seconds of exposure, a Hydrogen-alpha vertical transfer function was set to
with 120 seconds of exposure, and Blue 99.5% or 99.9%, based on trial and error
with 120 seconds of exposure for T27 to see which provided the clearest
and T30, Table 1. image. Each image was then measured
The luminance filter is a clear filter using the Distance and Angle tool in
that was used to get a general image of Mira Pro. This statistically located the
the system. Hydrogen-alpha filters were center of each star. Figure 3 is a CCD
used to even out the difference in image after being processed in
Maxim DL and MiraPro x64. Theta and
rho of the AD pair was recorded and
organized in an Excel spreadsheet.
Then, we determined the mean,
standard deviation, and the deviation of
error for all measurements.

Figure 1. Figure 2.
iTelescope T27. iTelescope T30.

Table 1. Number of images with filter and exposure time.

Figure 3. Processed CCD image of WDS 05599-3703B 1044B under a

luminance filter with an exposure of 60 seconds.
Data and Results

Table 2. The mean, standard error of the

mean, and standard deviation for WDS
05599-3703B 1044AD

Table 3. Summary of raw data from CCD

image measurements made in this study.
Table 4. Parallax and error measurements
for both stars

Table 5. Min, Mid, and Max distances from both stars to earth in
Parsecs and Light Years

Graph 1. Historical Position of WDS 05599-3703B


The conclusion that can be drawn from this data is that the system of stars is not truly
linked; their connection is simply due to optical alignment. There has been movement
observed, however, this movement is observed when the two stars cross over one
another. The large distance between these stars manifests that they are not connected.
To determine these distances, parallax angles were given, and these angles were used to
determine the distance of each star from earth in light years, coming out to be 173.4 light
years for star A and 1410.83 light years for star D, as seen in Table 5.

Table 6. Current measurements and data available on WDS 05599-3703B

1044AD courtesy of the Washington Double Star Catalog.
Stelledoppie Data
The online astronomical database,
Stelledoppie, states the nature of this
double is uncertain as of 1999.

The measurements from this study
indicate that there has been a decrease in
position angle and an increase in
separation for 1044AD since its first
recorded measurement in 1999. From
historical data and the data from this
study, it can be concluded that this is a
high proper motion double, however, the
stars are not gravitationally bound.

Genet, R., Johnson, J., Buchheim, R., and Harshaw, R. 2015, Small Telescope
Astronomical Research Handbook. Ed. Collins, D.
Mason, B. and Hartkopf, W. 2015, The Washington Double Star Catalog.
Astrometry Department, U.S. Naval Observatory.