Artificial Intelligence Diagnostic Approach of Infertility in Chinese Traditional Medicine
Abstract:
Bibliography/Citations:
Citations:
1. “Infertility.” World Health Organization, World Health Organization,
2. “5 Reasons Infertility Awareness Matters.” Carolinas Fertility Institute, 17 Apr. 2020,
3. “Traditional Chinese Medicine for Infertility: West Wimbledon Physio Clinic.” West Wimbledon Physiotherapy, 1 Feb. 2022,
4. Zhu, Jihe, et al. “Acupuncture Treatment for Fertility.” Open Access Macedonian Journal of Medical Sciences, Republic of Macedonia, 19 Sept. 2018,
5. Cyranoski, David. “Why Chinese Medicine Is Heading for Clinics around the World.” Nature News, Nature Publishing Group, 26 Sept. 2018,
6. Jiang, Lijuan. "Knowledge management system and infertility treatment using Traditional Chinese medicine." (2013).
7. Kumar, Yogesh, et al. “Artificial Intelligence in Disease Diagnosis: A Systematic Literature Review, Synthesizing Framework and Future Research Agenda.” Journal of Ambient Intelligence and Humanized Computing, Springer Berlin Heidelberg, 13 Jan. 2022
8. Xu, M.B., et al." Construction and application of TCM syndrome Differentiation model of infertility based on artificial intelligence." Chinese Journal of Traditional Chinese Medicine 36.09(2021):5532-5536.
9. Wang Y. Shi X. Li L. Efferth T. Shang D; “The Impact of Artificial Intelligence on Traditional Chinese Medicine.” The American Journal of Chinese Medicine, U.S. National Library of Medicine.
10. Wang, Yulin, et al. "The impact of artificial intelligence on traditional Chinese medicine." The American Journal of Chinese Medicine 49.06 (2021): 1297-1314.
11. Jordan M I, Mitchell T M. Machine learning: Trends, perspectives, and prospects[J]. Science, 2015, 349(6245): 255-260.
12. Arora, Surbhi. “Supervised vs Unsupervised vs Reinforcement.” AITUDE, 29 Jan. 2020,
13. Schober P, Vetter T R. Logistic regression in medical research[J]. Anesthesia and analgesia, 2021, 132(2): 365.
14. Tharwat A, Gaber T, Ibrahim A, et al. Linear discriminant analysis: A detailed tutorial[J]. AI communications, 2017, 30(2): 169-190.
15. Zhang Z. Introduction to machine learning: k-nearest neighbors[J]. Annals of translational medicine, 2016, 4(11): 218.
16. Pisner D A, Schnyer D M. Support vector machine[M]//Machine learning. Academic Press, 2020: 101-121.
17. Rana A, Rawat A S, Bijalwan A, et al. Application of multi-layer (perceptron) artificial neural network in the diagnosis system: a systematic review[C]//2018 International conference on research in intelligence and computing in engineering (RICE). IEEE, 2018:1-6.
18. Biau G, Scornet E. A random forest guided tour[J]. Test, 2016, 25(2): 197-227.
19. Dong X, Yu Z, Cao W, et al. A survey on ensemble learning[J]. Frontiers of Computer Science, 2020, 14(2): 241-258.
20. Song Y Y, Ying L U. Decision tree methods: applications for classification and prediction[J]. Shanghai archives of psychiatry, 2015, 27(2): 130.
21. Charbuty B, Abdulazeez A. Classification based on decision tree algorithm for machine learning[J]. Journal of Applied Science and Technology Trends, 2021, 2(01): 20-28.
22. Wong T T, Yeh P Y. Reliable accuracy estimates from k-fold cross validation[J]. IEEE Transactions on Knowledge and Data Engineering, 2019, 32(8): 1586-1594.
23. Beauxis-Aussalet E, Hardman L. Simplifying the visualization of confusion matrix[C]//26th Benelux Conference on Artificial Intelligence (BNAIC). 2014.
24. Shekar B H, Dagnew G. Grid search-based hyperparameter tuning and classification of microarray cancer data[C]//2019 second international conference on advanced computational and communication paradigms (ICACCP). IEEE, 2019: 1-8.
25. Kaur P, Khehra B S, Mavi E B S. Data augmentation for object detection: A review[C]//2021 IEEE International Midwest Symposium on Circuits and Systems (MWSCAS). IEEE, 2021: 537-543.
26. Shorten C, Khoshgoftaar T M, Furht B. Text data augmentation for deep learning[J]. Journal of big Data, 2021, 8(1): 1-34.
27. Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of artificial intelligence research, 2002, 16: 321-357.
Additional Project Information
Research Plan:
Rationale:
Infertility is a kind of reproductive disorder that frequently occurs among young couples. It is the failure to get pregnant after one year of sexual behavior. As it gradually drew more public attention, it was addressed infertility as nonnegligible since it induces severe social conflicts, like family violence, divorce, social stigma, emotional stress, depression, anxiety, and low self-esteem. According to Traditional Chinese Medicine (TCM) theory, infertility is mainly attributed to six inducements: Liver Qi Stagnation, Stagnation of Uterine Cells, Kidney Yang Deficiency, Kidney Yin Deficiency, Kidney Qi Deficiency, Internal Obstruction of Phlegm, and Dampness. From the TCM perspective, the disease can be diagnosed by observing a patient's body features like skin spots and feces. The effectiveness of TCM and systematic theoretical knowledge has been recognized widely. Yet, due to its complexity and massive work of diagnosis implying the underlying conceptual foundations, it is very challenging to diagnose disease manually to consider and weigh multiple factors simultaneously. Therefore, our research took advantage of Artificial Intelligence (AI) in medical use that could help us accurately dealing massive numerical data.
Research Aim and Introduction:
This topic focused on finding the relationship between syndrome and syndrome type of infertility and extracts features from many symptom data to predict syndrome type via an artificial intelligence approach. Therefore, we construct a model that most effectively diagnoses a patient's syndrome type of infertility using body features. The model should have practical application value and be potentially used in a clinical trial.
General Procedure:
1) Collect Data-We collected 600 valid cases from female infertility case records in the Gynecological Clinic of Traditional Chinese Medicine in China Rehabilitation Research Center Beijing Boai Hospital from October 2018 to March 2022 in the form of Excel.
2) Data Pre-processing-We deal with raw data using methods like missing value processing, classified data processing, feature selection, and extraction.
3) Program Designing-We designed six programs using six different artificial intelligence algorithms.
4) Model Training-We separate 600 individual samples into the training group and testing group in the ratio of 4:1 (480 samples for training models and 120 samples for testing models).
5) Analyze and Evaluate Results- We used a confusion matrix to derive the precision, recall, and weighted F1-score and compare these data of each algorithm to find the one that performed the best.
6) Model Adjust and Optimization- We further optimize the selected algorithm using Grid Search Hyperparameters and Data Augmentation methods like Synthetic Minority Oversampling Technique (SMOTE).
Bibliography:
“Infertility.” World Health Organization, World Health Organization,
“Traditional Chinese Medicine for Infertility: West Wimbledon Physio Clinic.” West Wimbledon Physiotherapy, 1 Feb. 2022,
Kumar, Yogesh, et al. “Artificial Intelligence in Disease Diagnosis: A Systematic Literature Review, Synthesizing Framework and Future Research Agenda.” Journal of Ambient Intelligence and Humanized Computing, Springer Berlin Heidelberg, 13 Jan. 2022
Xu, M.B., et al." Construction and application of TCM syndrome Differentiation model of infertility based on artificial intelligence." Chinese Journal of Traditional Chinese Medicine 36.09(2021):5532-5536.
Jordan M I, Mitchell T M. Machine learning: Trends, perspectives, and prospects[J]. Science, 2015, 349(6245): 255-260.
Schober P, Vetter T R. Logistic regression in medical research[J]. Anesthesia and analgesia, 2021, 132(2): 365.
Tharwat A, Gaber T, Ibrahim A, et al. Linear discriminant analysis: A detailed tutorial[J]. AI communications, 2017, 30(2): 169-190.
Zhang Z. Introduction to machine learning: k-nearest neighbors[J]. Annals of translational medicine, 2016, 4(11): 218.
Pisner D A, Schnyer D M. Support vector machine[M]//Machine learning. Academic Press, 2020: 101-121.
Rana A, Rawat A S, Bijalwan A, et al. Application of multi-layer (perceptron) artificial neural network in the diagnosis system: a systematic review[C]//2018 International conference on research in intelligence and computing in engineering (RICE). IEEE, 2018:1-6.
Biau G, Scornet E. A random forest guided tour[J]. Test, 2016, 25(2): 197-227.
Questions and Answers
1. What was the major objective of your project and what was your plan to achieve it?
a. Was that goal the result of any specific situation, experience, or problem you encountered?
The research aims to generate an optimized algorithm that diagnoses the infertility syndrome type. It is practical since 600 samples are applied for training the model, and over 50 characteristics are considered. The research has profound influences since infertility is broadly considered an inducing factor that causes severe social conflicts, like family violence, divorce, social stigma, emotional stress, depression, anxiety, and low self-esteem. Besides, applying such a method as an auxiliary is viable since the model eventually reached around 85% accuracy, and the number will increase as the model gets further trained.
b. Were you trying to solve a problem, answer a question, or test a hypothesis?
I was trying to solve the problem that provides a reliable prediction of infertility syndrome type and helps doctors apply appropriate treatment according to the syndrome type.
2. What were the major tasks you had to perform in order to complete your project?
a. For teams, describe what each member worked on.
I worked without forming any teams in this research. I finished most jobs, including plan forming, data source searching, coding, and essay editing. My instructor helped me to perfect the algorithm coding and mathematical model.
3. What is new or novel about your project?
a. Is there some aspect of your project's objective, or how you achieved it that you haven't done before?
My project is mainly innovative. People have brought up the idea of using artificial intelligence in medicine, including infertility diagnosis. However, it is the first time, as far as I know, that compare several models to test their efficiency in solving this problem. Besides, multiple methods are used to adjust the parameters to optimize accuracy. It may not be the first research about applying AI to the medical area. However, the thorough procedures for designing the best model are unprecedented. I didn’t know how to write an artificial program until I read several papers and books in this field, and my professor also instructed me about the programs.
b. Is your project's objective, or the way you implemented it, different from anything you have seen?
I implemented the project in a unique, scientific way. Since Artificial Intelligent methods may each be adapted to a different situation, and because this research aimed to provide help for human society, accuracy and efficiency are the two factors we mainly consider. Thus, we carefully compare and choose from the six algorithms, which is an innovative thought.
c. If you believe your work to be unique in some way, what research have you done to confirm that it is?
I’m sure this is a novel investigation because I haven’t seen any paper or research report that, similar to the methods I used to solve infertility, diagnosed the problem while searching for reliable article sources for reference.
4. What was the most challenging part of completing your project?
a. What problems did you encounter, and how did you overcome them?
I have encountered the problem during many procedures, including data collecting, method design, coding, result enhancement, and professional paper writing. However, the problem that could be more challenging for me is optimizing the result effectively. Before the professor instructed me, the adjusted parameter did not affect or improve the performance of the models. Soon after, I was introduced to Grid Searching Hyperparameters, which significantly increased my model’s efficiency.
b. What did you learn from overcoming these problems?
I learned much more professional knowledge about artificial intelligence and writing a scientific paper. More importantly, I learned to think thoroughly about problems and use the information provided online.
5. If you were going to do this project again, are there any things you would you do differently the next time?
I would say the whole research procedures are thorough and all reasonable, and the result from the testing group proved the usefulness of this program. However, I would probably search for more data samples if I could restart the program; much training data will provide more accurate parameters and models.
6. Did working on this project give you any ideas for other projects?
Absolutely. While working on this problem, I imagined we could apply artificial intelligence technology to other medical areas like the orthopedic and cardiovascular departments. Since bones and blood vessels are fragile parts of our body, we can only have limited information about the patient by looking at the X-ray film and B-scan ultrasonography. Since AI technology can often identify more details and are more experienced, they sometimes offer more reliable judgment than doctors.
7. How did COVID-19 affect the completion of your project?
COVID-19 has had limited influence on the completion of my project since I analyzed the model and was done writing papers at home. The professor and I often discuss it online through social media or at my house. The only issue that caused the COVID pandemic is that I didn’t get a chance to meet the doctor from whom I got the data source; I’m sure negotiating with her workgroup about my project will surely help me perfect my model further.