biomolecules

Editorial
Application of Artificial Intelligence for Medical Research
Ryuji Hamamoto 1,2

1 Division of Molecular Modification and Cancer Biology, National Cancer Center Research Institute,
5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan; rhamamot@ncc.go.jp
2 Cancer Translational Research Team, RIKEN Center for Advanced Intelligence Project, 1-4-1 Nihonbashi,
Chuo-ku, Tokyo 103-0027, Japan

The Human Genome Project, completed in 2003 by an international consortium, is

considered one of the most important achievements for mankind in the 21st century [1].
With the completion of this project, medical science has entered a new era known as
the post-genome era [2]. In the latter half of the 20th century and into the 21st century,
with the development of molecular biology, efforts to elucidate diseases at the molecular
level were actively pursued worldwide. Under these circumstances, the completion of the
whole human genome analysis has increased the momentum for personalized medicine,
i.e., patient-specific medicine optimized by combining genome information and molecular
medicine. Various technologies have been developed in the post-genome era; one of
the major technological advances is the advent of next-generation sequencing (NGS).
The duration of the human whole-genome analysis project undertaken by the international
consortium was 13 years, costing USD 3 billion to analyze the entire genome of a single
person [3]. However, with the advent of NGS, it now takes a day and less than USD 1000
for the same analysis [4]. Moreover, with the advent of such high-speed sequencers, the
amount of data obtained in medical research is enormous, and the term “big data” is now
common in medical research.
In the 21st century, with the progress of machine learning technology (especially the





 emergence of deep learning technology) and graphics processing units, big data analysis
using artificial intelligence (AI) technology is now common in various fields, including the
Citation: Hamamoto, R. Application
of Artificial Intelligence for Medical
medical field [5]. Its introduction in the medical field allows for a more objective analysis
Research. Biomolecules 2021, 11, 90.
of biological phenomena, which are inherently complex and diverse, and require careful
https://doi.org/10.3390/biom11010090 determination of the generalizability of the results obtained from analyses. In such an
academic field, if scientific discussions involve only limited data, then it becomes difficult
Received: 7 January 2021 to grasp the complete picture of a phenomenon, and it is easy to fall into a state of “you
Accepted: 11 January 2021 can’t see the forest for the trees.” In contrast, the analysis of large-scale data using AI
Published: 12 January 2021 technology will make it possible to elucidate biological phenomena more objectively and
without omission; it is expected to contribute greatly to the advancement of medicine.
Publisher’s Note: MDPI stays neu- In fact, more than 60 AI-powered medical devices are approved by the Food and Drug
tral with regard to jurisdictional clai- Administration (FDA) in the United States, and the use of AI in the medical field is trending
ms in published maps and institutio- worldwide [2].
nal affiliations. As shown in Figure 1, there are three areas in which AI technology is currently imple-
mented in the medical field: medical image analysis, omics analysis, and natural language
processing [2,5–9]. In this Special Issue on the “Application of Artificial Intelligence for
Copyright: © 2021 by the author. Li-
Medical Research”, articles were presented in the fields of medical image analysis and
censee MDPI, Basel, Switzerland.
omics analysis, among these three areas. As the editor of this Special Issue, I would like to
This article is an open access article
provide a brief overview of it.
distributed under the terms and con- Regarding radiation image analysis, Akatsuka et al. analyzed magnetic resonance
ditions of the Creative Commons At- images of the prostate with deep learning, compared them with observations by radiolo-
tribution (CC BY) license (https:// gists and pathologists, and showed that deep learning could identify cancerous areas at
creativecommons.org/licenses/by/ a high rate, and could also find useful clues for clinical diagnosis even when the cancer
4.0/). was not visible [10]. Sukegawa et al. confirmed that panoramic radiographs could be

Biomolecules 2021, 11, 90. https://doi.org/10.3390/biom11010090 https://www.mdpi.com/journal/biomolecules

Biomolecules 2021, 11, 90 2 of 4

analyzed using a deep convolutional neural network to accurately classify dental implant
systems [11]. Yamamoto et al. showed that hip radiographs could be analyzed using deep
learning to diagnose osteoporosis with high accuracy, and that adding clinical covariates
from patient records further improved performance [12]. For ultrasound image analy-
sis, Dozen et al. proposed a new segmentation method called Cropping–Segmentation–
Calibration, which is able to segment the ventricular septum in fetal ultrasound videos
using time-series information [13]. Kusunose et al. analyzed echocardiography images
with a convolutional neural network, and showed the possibility of classifying them into
five standard views (long axis, short axis, two-chamber view, three-chamber view, and
four-chamber view) [14]. As for skin image analysis, Jinnai et al. showed it is possible to
accurately classify malignant tumors (malignant melanoma and basal cell carcinoma) and
benign tumors (nevus, seborrhoeic keratosis, senile lentigo, and hematoma/hemangioma)
by training a convolutional neural network on images of pigmented skin lesions [15].

Figure 1. There are three areas in which AI technology is currently being actively implemented in the medical field:
medical image analysis, omics analysis, and natural language processing.

In this Special Issue, new directions in medical image analysis using AI technology
were also reported. Korb et al. measured serum samples using Fourier-Transform infrared
spectroscopy and analyzed the results using deep learning to show its potential in dis-
criminating between sera from healthy individuals, allergic patients, and patients treated
with allergen-specific immunotherapy [16]. Aida et al. used conditional generative adver-
sarial networks (CGANs) to segment cancer stem cells (CSCs) on phase-contrast images,
and showed the potential for mapping CSC morphology to an undifferentiated state using
a deep-learning CGAN workflow [17]. Kanada et al. proposed an efficient search method
for identifying successful regions by exploring the parameters of coarse-grained molecular
dynamic simulations using two machine learning methods: Bayesian optimization and
active learning [18]. Yamato et al. proposed a segmentation method using deep learning
Biomolecules 2021, 11, 90 3 of 4

to extract nerves from label-free endoscopic images obtained using coherent anti-Stokes
Raman scattering for nerve-sparing surgery [19].
In terms of omics analysis, Tanaka et al. used Bayesian networks to construct an
Epithelial–Mesenchymal Transition (EMT) network representing gene–gene interactions,
and showed that the sample-specific edge contribution value pattern of this EMT network
characterized the survival rate of lung cancer patients [20]. Using a publicly available
dataset (The Cancer Genome Atlas (TCGA) with a focus on lung adenocarcinoma (LUAD)),
Asada et al. succeeded in classifying good and poor prognosis groups by performing a
multi-omics analysis combining deep learning and machine learning, and also success-
fully identified genes contributing to the survival of LUAD patients [21]. Takahashi et al.
succeeded in classifying lung cancer patients based on their prognosis (poor or good) by
analyzing a multi-omics data set consisting of six categories of TCGA using a combination
of deep learning and machine learning [22]. Kobayashi et al. proposed a new method of
adding per-element input scaling to diet networks, and showed that lung cancer pathologi-
cal types (adenocarcinoma and squamous cell carcinoma) could be accurately identified
using somatic mutation profiles [23]. Ai et al. analyzed microarray gene expression data us-
ing weighted gene co-expression network analysis and a variational autoencoder to predict
colorectal cancer with high accuracy [24]. In their review, Lin et al. proposed the use of ma-
chine learning to improve the off-target properties of N-methylpyrrole-N-methylimidazole
polyamides (pyrrole-imidazole polyamides (PIPs)) [25].
It is clear that AI is a useful technology in the medical field, and I believe it is impera-
tive to keep abreast of the latest research findings to effectively utilize AI and overcome
challenges in the future. I sincerely hope that this Special Issue will be of some support
to readers.

Conflicts of Interest: The author declares no conflict of interest.

References
1. Collins, F.S.; Morgan, M.; Patrinos, A. The Human Genome Project: Lessons from large-scale biology. Science 2003, 300, 286–290.
[CrossRef] [PubMed]
2. Hamamoto, R.; Suvarna, K.; Yamada, M.; Kobayashi, K.; Shinkai, N.; Miyake, M.; Takahashi, M.; Jinnai, S.; Shimoyama, R.;
Sakai, A.; et al. Application of Artificial Intelligence Technology in Oncology: Towards the Establishment of Precision Medicine.
Cancers 2020, 12, 3532. [CrossRef] [PubMed]
3. Green, E.D.; Watson, J.D.; Collins, F.S. Human Genome Project: Twenty-five years of big biology. Nature 2015, 526, 29–31.
[CrossRef] [PubMed]
4. Levy, S.E.; Boone, B.E. Next-Generation Sequencing Strategies. Cold. Spring Harb. Perspect. Med. 2019, 9, a025791. [CrossRef]
5. Hamamoto, R.; Komatsu, M.; Takasawa, K.; Asada, K.; Kaneko, S. Epigenetics Analysis and Integrated Analysis of Multiomics
Data, Including Epigenetic Data, Using Artificial Intelligence in the Era of Precision Medicine. Biomolecules 2020, 10, 62. [CrossRef]
6. Komatsu, M.; Sakai, A.; Komatsu, R.; Matsuoka, R.; Yasutomi, S.; Shozu, K.; Dozen, A.; Machino, H.; Hidaka, H.; Arakaki, T.;
et al. Detection of Cardiac Structural Abnormalities in Fetal Ultrasound Videos Using Deep Learning. Appl. Sci. 2021, 11, 371.
[CrossRef]
7. Shozu, K.; Komatsu, M.; Sakai, A.; Komatsu, R.; Dozen, A.; Machino, H.; Yasutomi, S.; Arakaki, T.; Asada, K.; Kaneko, S.; et al.
Model-Agnostic Method for Thoracic Wall Segmentation in Fetal Ultrasound Videos. Biomolecules 2020, 10, 1691. [CrossRef]
8. Yamada, M.; Saito, Y.; Imaoka, H.; Saiko, M.; Yamada, S.; Kondo, H.; Takamaru, H.; Sakamoto, T.; Sese, J.; Kuchiba, A.; et al.
Development of a real-time endoscopic image diagnosis support system using deep learning technology in colonoscopy. Sci. Rep.
2019, 9, 14465. [CrossRef]
9. Hughes, K.S.; Zhou, J.; Bao, Y.; Singh, P.; Wang, J.; Yin, K. Natural language processing to facilitate breast cancer research and
management. Breast J. 2020, 26, 92–99. [CrossRef]
10. Akatsuka, J.; Yamamoto, Y.; Sekine, T.; Numata, Y.; Morikawa, H.; Tsutsumi, K.; Yanagi, M.; Endo, Y.; Takeda, H.; Hayashi, T.; et al.
Illuminating Clues of Cancer Buried in Prostate MR Image: Deep Learning and Expert Approaches. Biomolecules 2019, 9, 673.
[CrossRef]
11. Sukegawa, S.; Yoshii, K.; Hara, T.; Yamashita, K.; Nakano, K.; Yamamoto, N.; Nagatsuka, H.; Furuki, Y. Deep Neural Networks
for Dental Implant System Classification. Biomolecules 2020, 10, 984. [CrossRef] [PubMed]
12. Yamamoto, N.; Sukegawa, S.; Kitamura, A.; Goto, R.; Noda, T.; Nakano, K.; Takabatake, K.; Kawai, H.; Nagatsuka, H.;
Kawasaki, K.; et al. Deep Learning for Osteoporosis Classification Using Hip Radiographs and Patient Clinical Covariates.
Biomolecules 2020, 10, 1534. [CrossRef] [PubMed]
Biomolecules 2021, 11, 90 4 of 4

13. Dozen, A.; Komatsu, M.; Sakai, A.; Komatsu, R.; Shozu, K.; Machino, H.; Yasutomi, S.; Arakaki, T.; Asada, K.; Kaneko, S.; et al.
Image Segmentation of the Ventricular Septum in Fetal Cardiac Ultrasound Videos Based on Deep Learning Using Time-Series
Information. Biomolecules 2020, 10, 1526. [CrossRef] [PubMed]
14. Kusunose, K.; Haga, A.; Inoue, M.; Fukuda, D.; Yamada, H.; Sata, M. Clinically Feasible and Accurate View Classification of
Echocardiographic Images Using Deep Learning. Biomolecules 2020, 10, 665. [CrossRef]
15. Jinnai, S.; Yamazaki, N.; Hirano, Y.; Sugawara, Y.; Ohe, Y.; Hamamoto, R. The Development of a Skin Cancer Classification System
for Pigmented Skin Lesions Using Deep Learning. Biomolecules 2020, 10, 1123. [CrossRef]
16. Korb, E.; Bagcioglu, M.; Garner-Spitzer, E.; Wiedermann, U.; Ehling-Schulz, M.; Schabussova, I. Machine Learning-Empowered
FTIR Spectroscopy Serum Analysis Stratifies Healthy, Allergic, and SIT-Treated Mice and Humans. Biomolecules 2020, 10, 1058.
[CrossRef]
17. Aida, S.; Okugawa, J.; Fujisaka, S.; Kasai, T.; Kameda, H.; Sugiyama, T. Deep Learning of Cancer Stem Cell Morphology Using
Conditional Generative Adversarial Networks. Biomolecules 2020, 10, 931. [CrossRef]
18. Kanada, R.; Tokuhisa, A.; Tsuda, K.; Okuno, Y.; Terayama, K. Exploring Successful Parameter Region for Coarse-Grained
Simulation of Biomolecules by Bayesian Optimization and Active Learning. Biomolecules 2020, 10, 482. [CrossRef]
19. Yamato, N.; Matsuya, M.; Niioka, H.; Miyake, J.; Hashimoto, M. Nerve Segmentation with Deep Learning from Label-Free
Endoscopic Images Obtained Using Coherent Anti-Stokes Raman Scattering. Biomolecules 2020, 10, 1012. [CrossRef]
20. Tanaka, Y.; Tamada, Y.; Ikeguchi, M.; Yamashita, F.; Okuno, Y. System-Based Differential Gene Network Analysis for Characteriz-
ing a Sample-Specific Subnetwork. Biomolecules 2020, 10, 306. [CrossRef]
21. Asada, K.; Kobayashi, K.; Joutard, S.; Tubaki, M.; Takahashi, S.; Takasawa, K.; Komatsu, M.; Kaneko, S.; Sese, J.; Hamamoto, R.
Uncovering Prognosis-Related Genes and Pathways by Multi-Omics Analysis in Lung Cancer. Biomolecules 2020, 10, 524.
[CrossRef] [PubMed]
22. Takahashi, S.; Asada, K.; Takasawa, K.; Shimoyama, R.; Sakai, A.; Bolatkan, A.; Shinkai, N.; Kobayashi, K.; Komatsu, M.;
Kaneko, S.; et al. Predicting Deep Learning Based Multi-Omics Parallel Integration Survival Subtypes in Lung Cancer Using
Reverse Phase Protein Array Data. Biomolecules 2020, 10, 1460. [CrossRef] [PubMed]
23. Kobayashi, K.; Bolatkan, A.; Shiina, S.; Hamamoto, R. Fully-Connected Neural Networks with Reduced Parameterization for
Predicting Histological Types of Lung Cancer from Somatic Mutations. Biomolecules 2020, 10, 1249. [CrossRef] [PubMed]
24. Ai, D.; Wang, Y.; Li, X.; Pan, H. Colorectal Cancer Prediction Based on Weighted Gene Co-Expression Network Analysis and
Variational Auto-Encoder. Biomolecules 2020, 10, 1207. [CrossRef] [PubMed]
25. Lin, J.; Nagase, H. The Road Not Taken with Pyrrole-Imidazole Polyamides: Off-Target Effects and Genomic Binding. Biomolecules
2020, 10, 544. [CrossRef] [PubMed]