PRE2025 1 Group1

Name	Student number	Study	E-mail
Robert Arnhold	1847848	Mechanical Engineering	r.w.arnhold@student.tue.nl
Sietse Bosman
Octavian Astefanei	1836374	Electrical Engineering	o.astefanei@student.tue.nl
Anne Willems	1631810	Electrical Engineering	a.m.j.e.willems@student.tue.nl
Kerim Gjergjizi	1813420	Electrical Engineering	k.gjergjizi@student.tue.nl
Lucas Spronk	1563564	Computer Science	l.spronk@student.tue.nl

Plan

Concept Introduction

This project aims to design and prototype a smart hamster cage. Hamster cages have long featured similar designs, features and limitations. While a hamster wheel may give the hamster the ability to exercise, a water bottle give it the ability to drink and a food bowl the ability to eat, these are all parts of the hamster cage that are long outdated. In our modern world there is a market gap: a hamster cage that can help care for your hamster, teach you more about it, and let you better interact with your beloved pet. If successful, this proof-of-concept opens the door to an expansion towards larger pet enclosures such as rabbits, guinea pigs, and more.

Objectives

• Provide reliable care for the hamster, even while users are away
• Help users more effectively care for their hamsters
• Other, user-defined objectives that will be found via user research

Users

General Target Audience:

The target audience consist of hamster owners themselves, as well as anyone else involved in caring for the hamster, e.g. owner's family or roommates.

People of various ages own hamsters, but they are mostly "millennials, aged 25-39", who "make up about one in three small-pet owners" (also includes fish, birds and some other animals) [1]. There is a strong correlation with younger aged owners, especially the presence of household members under the age of 12; "approximately 72% of hamster owns are under the age of 44" [2], and "nearly 90% of households with hamsters include children, and 87% of those households have children under 12 years old" [3].

While small-pet owners in the US tend to earn lower incomes than other pet owners [2], they tend to spend more on their pets. With a significant margin too, averaging about 252USD/month as opposed to the next largest, dog owners, with around 140USD/month. Furthermore, a general trend in all pet owners was found when comparing spending by gender and relationship status; women spent 117USD/month on their pets compared to men spending 137USD/month, and married owners spent 132USD/month as opposed to single owners spending 129USD/month and unmarried partners with 124USD/month. Finally, users most willing to spend on their pets were led by the 18-24 age range with 174USD/month, followed by 25-34 with 142USD/month, 35-44 with 108USD/month, and further decreasing with age. Noteworthy is, however, that the age group with the lowest average spending is ages 14-17 with 73USD/month [4]. While there is a general willingness to spend upwards of 100USD on a pet every month, the final product should aim to appeal to users of all ages, including allowing young owners to buy the hamster cage themselves without being a burden to their parents.

Finally, owners of small pets are most commonly "urban dwellers, particularly those [living] in apartments or smaller living spaces". They "disproportionately own compact pets due to space constraints" [5], and since hamsters are quite restricted in their spacial requirements compared to other larger animals or animals with special living conditions, hamsters are very popular in this demographic category. However, the majority of families with children live in

To conclude, a preliminary target market can be narrowed down to users between 12 and 44 years old, typically either children or their parents, who either care for the hamster themselves or live with family who helps take care of the pet. Users will typically be willing to spend upwards of 73 to 137USD/month, which suggests a further increased price range for a one-time purchase.

Example User Profiles:

Emma, 12 years old (student)

• Recently got her first hamster as a pet.
• Parents want her to learn responsibility but also want to ensure pet safety.
• Would benefit from automated feeding/water alerts and activity data.
• Likes fun graphics showing hamster’s daily life.

Michael, 28 years old (busy professional)

• Works long hours and sometimes travels for a few days.
• Wants peace of mind that his hamster has food/water when away.
• Needs reliable notifications and refill reminders.
• Less interested in visuals, more in core “care assurance” features.

Sofia, 38 years old (parent of two children)

• Sarah has two children, ages 7 and 10.
• Children are responsible for daily care but Sarah supervises to ensure the hamster’s well-being.
• Values tools that make pet care educational and fun.
• Wants quick access to reliable information to monitor the hamster throughout the day.

Approach

Potential Features

• A camera needs to be added to the hamster cage, which such that the owner can monitor the hamster off site.
• A automatic feeder needs to be added, which adheres to a healthy feeding pattern. In a way that the availaility of the user is not required for the feeding.
• A measurement system for the food and water intake needs to be added. This to spot abnormalities in the eating pattern of the hamster.

Literature studies

A low-cost automated apparatus for investigating the effects of social defeat in Syrian hamsters [7]

Relevant points/summary: The article is about an automated apparatus that can used to investigate the effects of defeat in hamsters. They use low power lasers and laser detectors to keep track of the position of the master. A computer measures three things using the lasers\detectors:  average position, frequency of visits to each chamberand the duration of these visits ,and the frequency of changes in direction of travel in each chamber. The data collection program they use is MEDPC for windows.

An Open-Source, Automated Home-Cage Sipper Device for Monitoring Liquid Ingestive Behavior in Rodents [8] 

Relevant points/summary:

Common approaches for measuring liquid intake:

-       Computer-tethered lickometers

-       Video based systems

-       Measuring or weighing the liquid

In the article they made a photobeam-based sipper device.

Advantages:

- battery powered (battery life of > 2 weeks)

- fits in vivarium caging

- quantifies the intake of two different liquids simultaneously

-  low cost and easily constructed

- provides data with high temporal resolution (allows for detailed analysis of drinking patterns)

- open source

Disadvantage

-       Animal interactions that are picked up by the sensor are not limited to licks

In the article they use the device to measure the volume of liquid ingested, whether there is a preference for chocolate milk vs water and confirm that the sipper device can be succesfuly integrated with in vivo measurements.

A System for Monitoring Animals Based on Behavioral Information and Internal State Information [6]

     Relevant points/summary:

     In this paper, a monitoring system for animals using video image analysis is used. It  extracts features related to behavorial information and the animal internal state via mask R-CNN. These features are used to detect typical daily activities and anomalous activities. This way it can detected when the hamster behaves in an unusual way. The system also combines a new feature extraction method using deep neural network techology and an anomaly detection method.

The systems consists of 4 parts:

-       Signal measurement: a digital hi-vision video camera

-       Feature extraction: mask R-CNN and color information detection

-       State discrimination: one class SVM

-       Display: PC with an HDMI capture device

To test the system a loud sound was generated and the hamsters reaction was measured.

The discrimination rate of non-daily activities did not have a high level of accuracy. Mainly because the hamster shrank during the daily activities and the direction vector was often inverted because of the noise caused by the hamster’s shadow. The training data also includes infrequent situations in daily activities, which makes it possible that the discriminative boundary was not accurately determined

Design and Development of a Smart Pet Feeder with IoT and Deep Learning [9]

Relevant points/summary:

Automatic pet feeder was developed using internet of things technolgy and deep learning to address feeding challenges. The system makes sure the pet receives the right amount of food, regardless of whether the owner is available. This makes sure the animal stays healthy, because an owners busy schedule can often lead to inconsistent feeding, which can cause malnutrition or obesity. It is also offers pet owners convenience and peace of mind by enabling the feeding process in their absence. 

The system architecture:

-       Weight sensor (Tecneu HX711): determines pets weight and food weight

-       Camera: detects which species

-       Ultrasonic sensor (HC-SR04): measures distance to the pet from the feeder.

-       Servo motor: controls dispensing mechanism

-       Arduino (mega 2560): microcontroller that handles all the actuators and sensors.

Areas of improvement that were mentioned:

The system could learn from a pets’ eating patterns and adjust feeding schedules and portion sizes over time and adding other pet identification, since only dogs were used to train this system.

A Novel Automated System Yields Reproducible Temporal Feeding Patterns in Laboratory Rodents [10]

Relevant points/summary:

In this article the aims was to develop and validate a reliable method for supplying crushed diets to laboratory rodents in consistent, relevant feeding patterns for prolonged periods.

They used 2 different feeding patterns:

-nocturnal meal-feeding

-nocturnal grazing

The cumulative food intake for both methods was the same.

PiE: an open-source pipeline for home cage behavioral analysis [11]

Relevant points/summary:

In this article PiE is introduced, an open-source, end-to-end, user configurable, scalable, and inexpensive behavior assay system. It has a custom-built behavior box to hold a home cage, as well as software enabling continuous video recording and individual behavior box environmental control. The PiE system reduces environmental and experimenter disruptions by providing fully remote control and monitoring of home cage behaviors. The system was tested on mice. The sensors and actuators that were used were  a ceiling mounted, downward facing, infrared (IR) video camera (Pi NoIR) to allow both daytime and nighttime video recording, (2) white and IR LEDs to illuminate the behavior box (Note: to match the Pi NoIR camera sensor, 940nm or shorter wavelength IR LEDs are preferred for optimal nighttime illumination), (3) a temperature and humidity sensor, and (4) a circulating fan for climate stability, which were controlled by a raspberry pi. Desktop software, called VideoAnnotate, was used to perform behavioral analysis.

CageView: A Smart Food Control and Monitoring System for Phenotypical Research In Vivo [12]

Relevant points/summary:

The article introduces an automated and smart system (named CageView) used to monitor a mouse, detect motion, and control access to food in accordance with experimental schedules. Cageview accomplishes it food control and activity monitoring via:

1. An actuator for linear displacements of the food access door controlled by a custom-designed interface to set the feeding and fasting schedule;

2. A vision unit with visible and near-infrared cameras and a near mid-infrared LED for day and nighttime monitoring, which forms a video streaming and data transmission system using wireless or wired communication networks; It also uses a a Raspberry Pi single-board computer.

3. A trained convolutional neural network (CNN) that detects the animal position in the image which enables movement measurement.

The feeding mechanism is designed to be mounted on conventional cages used in most research centers and has a sliding mechanism that can enable or disable the animal’s access to food. Activity measurement is done via making an activity heatmap and distance measurement.

The CageView technology has been disclosed in the Maddahi Y. and Maddahi A. Methods and apparatus for monitoring, feeding, and checking animals. United States Patent. US 63/321,368, 2022

Feeding Experimentation Device (FED): Construction and Validation of an Open-source Device for Measuring Food Intake in Rodents [13]

Relevant points/summary:

In the article they provide a description of a solution for measuring food intake by mice,  FED. FED is an open-source system that was designed to facilitate flexibility in food intake studies. It is compact and battery powered so it can fit in standard home cages.  The mice are fed using food pellets. When a mouse removes a pellet, a photointerrupter sensor sends a signal to the microcontroller and the time-stamp is logged on the onboard secure digital (SD) card. A new pellet is dispensed via the motor. The feeding can be limited to specific times of the day. The limitations of the FED:

-       SD card can be a cumbersome means to track and store data from many FEDs

-       Pellet jams could occur

-       Users should never leave mice with FED as their only food source without checking FED's functionality daily

A single FED can be assembled for approximately $350.

An open source automatic feeder for animal experiments [14]

Relevant points/summary:

In the article they describe an open source experimental feeder using an Arduino microcontroller. This feeder can be used for most sizes of dry food pellets with the potential modification of a single component;  They also describe the buidling procedure of the feeder. They tested it on pigeons, monkeys and cats. The cost for building this feeder is less than 200 euros.

The Promise of Automated Home-Cage Monitoring in Improving Translational Utility of Psychiatric Research in Rodents [15]

Relevant points/summary:

In the article they describe three of the most commonly used approaches for automated home cage monitoring in rodents and review several commercially available systems that integrate the different approaches. Automated home-cage monitoring records rodent behavior in their home cage over extended periods using minimal human contact.

Approaches:

-       Operant wall systems: detect nose pokes using an infrared beam and deliver food pellets or liquid in response to learned behaviors. Behaviors are directed by light or tone cues controlled by the operant wall.

-       computerized visual systems: CVS monitor behavior at high temporal resolution over extended periods including the dark and light phases of the circadian cycle, and trainable software are publicly available. The trainable nature of CVS allows for extensive flexibility in terms of the behavioral phenotypes it monitors. Limitations of the CVS include the requirement for adequate contrast between the rodent and its background and the need for most systems to house animals individually

-       Automatic motion sensors: use motion sensors to assess locomotor activity.  AMS generate data about distance traveled, velocity, and time spent at specific locations within the cage. They are unobtrusive and are not affected by lighting conditions, allowing for reliable data collection during the dark cycle on locomotor activity and sleep behavior

Bibliography

• [1] Packaged Facts. (2019, February 13). Baby boomers and millennials are redefining modern pet ownership trends, reports Packaged Facts. Business Insider. https://markets.businessinsider.com/news/stocks/baby-boomers-and-millennials-are-redefining-modern-pet-ownership-trends-reports-packaged-facts-1028733944
• [2] Zoonerdy. (2022, October 19). What is the number of hamsters in the United States? Zoonerdy. https://zoonerdy.com/what-is-the-number-of-hamsters-in-the-united-states
• [3] PR Newswire. (2013, February 12). Pet population and pet owner trends in the U.S.: Fish, birds, reptiles, and small animals. PR Newswire. https://www.prnewswire.com/news-releases/pet-population-and-pet-owner-trends-in-the-us-fish-birds-reptiles-and-small-animals-191262301.html
• [4] Petfood Industry. (2022, July 27). Survey examines U.S. pet ownership demographics. Petfood Industry. https://www.petfoodindustry.com/pet-food-market/article/15464848/survey-examines-us-pet-ownership-demographics
• [5] P Market Research. (2023). Household small animal treats market. P Market Research. https://pmarketresearch.com/hc/household-small-animal-treats-market/
• [6] T. Shibanoki, Y. Yamazaki, H. Tonooka (2024). A System for Monitoring Animals Based on Behavioral Information and Internal State Information. Animals. https://www.mdpi.com/2076-2615/14/2/281
• [7] Askew, A., González, F. (2014) A low-cost automated apparatus for investigating the effects of social defeat in Syrian hamsters. Behav Res 46, 1013–1022 . https://doi.org/10.3758/s13428-013-0427-x
• [8] E. Godynyuk, M.N. Bluitt, J.R. Tooley, A.V. Kravitz, M.C. Creed (2019) An Open-Source, Automated Home-Cage Sipper Device for Monitoring Liquid Ingestive Behavior in Rodents https://pmc.ncbi.nlm.nih.gov/articles/PMC678734
• [9] O.E. Castillo-Arceo, R.U. Renteira-Flores, P.C. Santana-Mancilla (2024) Design and Development of a Smart Pet Feeder with IoT and Deep Learning https://www.mdpi.com/2673-4591/82/1/63
• [10] T.W. Tilston, R.D. Brown, M.J. Wateridge, B. Arms-Williams, J.J. Walker, Y. Sun, T. Wells (2019) A Novel Automated System Yields Reproducible Temporal Feeding Patterns in Laboratory Rodents. Elsevier. https://www.sciencedirect.com/science/article/pii/S0022316622167295?via%3Dihub
• [11] J. Benedict, R.H. Cudmore (2023) PiE: an open-source pipeline for home cage behavioral analysis. Neurosci https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2023.1222644/full
• [12] M. Saeedi, A. Maddahi, A.M. Nassiri, M. K. Zareina (2022) CageView: A Smart Food Control and Monitoring System for Phenotypical Research In Vivo. applied sciences. https://www.mdpi.com//2076-3417/12/10/4966
• [13]K.P. Nguyen, M.A. Ali, T.J. O'Neal, I. Szczot, J.A. Licholai, A.V. Kravitz (2017) Feeding Experimentation Device (FED): Construction and Validation of an Open-source Device for Measuring Food Intake in Rodents. J Vis Exp. https://pmc.ncbi.nlm.nih.govarticles/PMC5409291//
• [14] J. Oh, R. Hofer, W.T. Fitch (2016) An open source automatic feeder for animal experiments. HardwareX.https://www.researchgate.net/publication/309307810_An_open_source_automatic_feeder_for_animal_experiments
• [15] A. Mingrone, A. Kaffman, A. Kaffman (2020) The Promise of Automated Home-Cage Monitoring in Improving Translational Utility of Psychiatric Research in Rodents. frontiers in neuroscience. https://pmc.ncbi.nlm.nih.gov/articles/PMC7773806/

Weekly Tasks

Week 1

Name	Total Time Spent	Task Breakdown
Robert Arnhold
Sietse Bosman
Octavian Astefanei
Anne Willems
Kerim Gjergjizi
Lucas Spronk

Week 2

Name	Total Time Spent	Task Breakdown
Robert Arnhold
Sietse Bosman
Octavian Astefanei
Anne Willems
Kerim Gjergjizi
Lucas Spronk