The Most Cited Articles in Neuroimaging for Depression: A Bibliometric Analysis of the Top 100 Most Highly Cited Articles Between 1992 and 2020

Mustafa Salimeen^1*, Abdoulaye Issotina Zibrila² and Mazen Musa³

¹Department of Radiology, Dongola Teaching Hospital, Faculty of Medicine and Health Sciences, University of Dongola, Dongola, Republic of Sudan, Dongola, 11111.

²Laboratory of Experimental Pharmacology, Department of Animal Physiology, Faculty of Science and Technology, University of Abomey Calavi, Benin

³Department of Orthodontics, Al Tegana Dental Teaching Hospital, Faculty of Dentistry, University of Science and Technology, Omdurman 11111, Khartoum, Sudan

^*Corresponding Author: Radiology Department, Dongola Teaching Hospital, Fuculty of Medicine and Health Sciences, University of Dongola, Sudan.

Received: 25 September 2022; Accepted: 03 October 2022; Published: 14 October 2022

Introduction

Major depression or depression is a common illness that severely limits psychosocial functioning and diminishes the quality of life around the world [1]. In 2008, the World Health Organization (WHO) ranked major depression as the common cause of death worldwide and projected that the disease will rank first by 2030.[2] The global prevalence of major depressive illness is estimated at around 4.7 %, with a 3% yearly incidence rate.[3] Depression affects 50% of all psychiatric outpatients and 12% of all inpatients, significantly influencing the quality of life.[4] Every year, about 5.8% of males and 9.5 % of females experience depression episodes [5]. According to the WHO, depression is the fourth major cause of disability worldwide [6]. With thousands of studies published, neuroimaging has become one of the most fundamental approaches to understanding depression. On the other hand, quantitative studies on neuroimaging in depressive patients are still limited. It is challenging for medical students, junior residents, and new researchers to determine essential study areas and future directions in this profession.

Bibliometric methods have widely been utilized to evaluate the impact of research outputs and analyze books and papers [7]. This form of research identifies the countries, institutions, and authors who have made the most significant contributions to science [8]. Highly cited papers' subjects, study designs, and levels of evidence-based medicine may affect clinical practice and future research [9]. The citation rate of an article usually correlates with the researchers' desire to use such a paper in their research. As a result, bibliometric analysis can describe the current state of a disease or study subject and provide ideas and recommendations for future research [10].

Very few medical majors have used rank analysis to find the most significant publications in their field.[11-14] There have been no studies to select the most influential papers on neuroimaging in depression. As a result, we used a bibliometric analysis approach to quantitatively sort out the knowledge system in the field of depression in neuroimaging by analyzing each article individually for aspects such as publication year, affiliation and country of the first author, study design, study purpose, citation count, yearly citation, imaging modality, journals, and journals' impact factor (IF).

Materials and Methods

Our research retrospective analyzes already published literature exempted from institutional review board approval.

Data Acquisition

All included publications were retrieved and obtained from the Web of Science Core Collection (https://www.webofscience.com/wos/woscc/ advanced-search) (Clarivate 30 Thomson Place, 36T3 Boston, DE 02210) in this study on December 5, 2021. The true findings were limited to English language research articles. The search terms were listed: 1# TS= (("depression" or "major depression disorders" or "MDD" or "major depression" or "anxiety" or "anxiety depression" or "vascular depression" or "mood disorders" or "bipolar disorder" or "unipolar disorder" or "unipolar depression" or "UD" or "bipolar depression" or "BD") 2# TS= ("neuroimaging" or "neuroradiology" or "brain imaging" or "computed tomography" or "brain CT" or "head CT" or "magnetic resonance imaging" or "MR imaging" or "MRI" or "functional magnetic resonance imaging" or "fMRI" or "structural magnetic resonance imaging" or "sMRI" or "diffusional kurtosis imaging" or "DKI" or "diffusion-weighted imaging" or "DWI" or "diffusion tensor imaging" or "DTI" or "MR perfusion" or "magnetic resonance perfusion" or "PWI" or "perfusion weighted imaging" or "ASL" or "arterial spin labelling" or "dynamic susceptibility contrast enhanced" or "DSC "or "magnetic resonance spectroscopy" or "MRS" or "positron emission tomography" or "PET" or "single photon emission computed tomography" or "SPECT" or "sonography" or "network" or "connect" or "sonography" or "ultrasound" or "Doppler" and "infant" or "child" or "pediatric" or "toddler" or "baby" or "kid" or "neonate" or "newborn" or "adolescent" or "teenager" or "juvenile" or "teen" or "adult" or "Offspring")). Indexes = Science Citation Index Expanded (SCIE), Social Sciences Citation Index (SSCI), and Arts and Humanities Citation Index (A&HCI) from the Web of Science Core collection. Sources of Emergence Proceedings Citation Index Science; ESCI; Conference Proceedings Citation Index Social Science (CPCI-S) & Humanities (CPCI-SSHI); Science Book Citation Index (BKCI-S); Book Citation Index for the Social Sciences and Humanities (BKCI-SSH); timespan = 1992–2021, 1# AND 2#. Articles that took advantage of those neuroimaging modalities to explore the diagnosis, management outcomes and mechanism of depression were included. Conference abstracts, case reports, and nonhuman studies were excluded.

Screening of the Most Highly Cited Original Research Articles

We utilized the included original articles for the sake of bibliometric analysis. Numerous variations distinguished the names of the major authors who appear in the report. The findings revealed annual publication trends. It was also checked for references and institutions. The current neuroimaging management for depression was put to the test through analysis of visualization. In descending order of yearly citations (the total citation count divided by the year difference between publication year and 2021), the top 100 articles were chosen.

Data Analysis

The included utilized articles were extended between 1992–2020; for the most highly cited literature included in the final study analysis, the following data were collected: publication year, contributing affiliation and country of the first author, study design, study purposes, total citations, yearly citations, imaging modality, publication journal and journal's IF. For descriptive analysis, the numbers and proportion of the literature were provided.

Results

Among the 182,551 published articles, 5153 were identified based on the inclusion and exclusion criteria. Based on their ranking number of yearly citations, 100 articles were recorded as the most highly cited articles. For collaboration analyses, 501 original articles were obtained. The number of studies published each year in the last four decades is listed in (table 1). Although the count of studies fluctuated from 1992 to 2020, the overall trend remained and ranked.

The Top 100 Most Highly Cited Articles

The top 100 most highly cited original research articles are revealed in Table 1. The yearly citation counts ranged from 14.29 to 147.00 per year (median: 80.67). The total citation counts were between 19 and 1883 (median: 951 times). Based on different study purposes, the articles were sorted out into three subdivisions: studies related to the mechanisms of depression (78 articles, 78%), those related to the prognosis of depression (13 articles, 13 %), and those related to the diagnosis of depression (9 articles, 9 %). According to the classification, we analyzed the publication year, contributing affiliation and country of the first author, study design, study purposes, total citations, yearly citations, imaging modality, publishing journal, and journal's IF of these articles

Table 1: The Top 100 Most Highly Cited Articles in Depression Ranked by Total Citation

Note: sMRI, structural magnetic resonance imaging; fMRI-RS, functional magnetic resonance imaging resting-state; DTI, diffusion tensor imaging; SPECT, single-photon emission computed tomography; PET, positron emission tomography, fMRI-TB= functional magnetic resonance imaging tract-based; fMRI-BOLD=functional magnetic resonance imaging blood-oxygen-level-dependent; fMRI-ER= functional magnetic resonance imaging event related, MRS, magnetic resonance spectroscopy

Study Designs of the Top 100 Most Highly Cited Articles

Table 2 demonstrates the article counts, study designs, as well as publication years and their distributions among different study aims and medians. Depression case-control studies contained three (03) diagnosis studies, 51 mechanism-related studies, and four (04) prognosis and outcome studies which were reported from 1992-2019 (median: 2005.5). Cross-sectional studies on depression were composed of two (02) diagnoses, 18 mechanism-related, and five (05) prognosis and outcome studies which were reported from 1997-2020 (median: 2008.5). Randomized controlled trial studies on depression consisted of three (03) diagnoses, one (01) mechanism-related study, and three prognosis and outcome studies and were reported from 2000-2017(median: 2008.5). Multicenter studies related to depression had one (01) diagnosis and four (04) mechanism-related studies (2005-2020, median: 2017). Clinical trial studies related to depression are divided into four (04) mechanism-related studies and one (01) focused on prognosis and outcome studies (1998-2010, median: 2004). For mechanism studies (1992-2020, median: 2006), functional magnetic resonance image (fMRI) assumed an enormous and more effective modality. In addition, subcallosal cingulate gyrus deep brain stimulation received attention for treatment-resistant depression, especially the positron emission tomography (PET) studies. The prognosis evaluating studies (1997-2020, median: 2017) mainly focused on assessing the therapeutic effect through different neuroimaging modalities.

Table 2: Depression Study Designs of Highly Cited Articles in Different Study Topics

Study Designs

The sample sizes in the studies ranged from 10 to 6503, with median and mean sample sizes of 47.5 and 223.52, respectively. Among the top 100 most highly cited articles, 58% (58/100) were the case-control, and 25% (25/100) were cross-sectional studies. Most depression-related studies used one of these two study designs. Meanwhile, a prospective design was employed in a randomized control trial 7% (7/100) were three (03) diagnoses, one (01) mechanism, and three (03) therapeutic outcome related studies. For multicenter study representing 5% (5/100), there was one (01) diagnosis and four (04) mechanism-related studies. As for clinical trial studies occupying 5% (5/100) of all studies, there were four (04) mechanisms and one (01) therapeutic outcome-related study. The diagnosis studies are nine (09) included three (03) case-control (3%), two (02) cross-sectional studies (2%), three (03) randomized control trial studies (3%), one (01) multicenter study (1%).

Application of Imaging Modalities in the Top 100 Highly Cited Articles

Figure 1A illustrates the annual publication count distribution of the top 100 highly cited original publications that used single or multiple modalities. Between the years (2009-2015), only four (04) studies were multi-modality studies focused on depression mechanism, while ninety-six studies used single modality (depression diagnosis: n = 09; depression mechanism: n = 74; depression prognosis and therapeutic effect: n = 12). Multiple imaging modalities were used in four (04) of the multi-modality studies. As a result, a single imaging modality could be enough to understand different mechanisms or get a proper diagnosis for depression brain imaging studies.

Figure 1B, according to distinct study aims, presents the yearly article counts of each imaging modality employed in the top 100 highly cited original articles. First, PET (1992) was the first method utilized in depression mechanism research (n = 17), fMRI was the most imaging modality used (45 times), DTI (09 times), and sMRI (31 times). Second, between 2009 and 2015 (4/100 research), multi-modalities were used less frequently in depression mechanism studies, particularly DTI combined with other MR technologies, sMRI and fMRI (sMRI & fMRI) together in one study, and MRS with fMRI.

Figure 1: The Distribution of Imaging Modality and Study Topics. (A) Yearly total count distribution and increasing trend of top 100 highly cited articles using single or multi-modality. (B) The number of articles per imaging modality in the top 100 highly cited articles about depression diagnosis studies, depression mechanism studies, and depression prognosis studies. Different topic studies are displayed in different colors and show different application trends within years. sMRI: Structural magnetic resonance imaging; DTI: Diffusion tensor imaging; fMRI-RS: Functional magnetic resonance imaging resting-state; fMRI-TB: Functional magnetic resonance imaging tract-based; fMRI-BOLD: Functional magnetic resonance imaging blood-oxygen-level-dependent; fMRI-ER: functional magnetic resonance imaging event-related; SPECT: Single-Photon Emission Computed Tomography; PET: Positron Emission Tomography; MRS: Magnetic resonance spectroscopy.

Authors and Journals of the Top 100 Most Highly Cited Articles

Authors' affiliation

The countries of affiliation of the first author are listed in descending order as follows: United States of America (54, 54%), United Kingdom (10, 10%), Germany (7, 7%), China (6, 6%), Netherlands (5, 5%), Canada (4, 4%), Ireland (3, 3%) France (2, 2%), Australia (2, 2%), Italy (2, 2%), Switzerland (2, 2%), Singapore (2, 2%), Norway (1, 1%).

Among the first authors, most were affiliated with the psychiatry and behavioral sciences (52/100, 52%). A minority of the authors were affiliated with neuropsychology and mental health (14/100, 14%), department of neurosciences (11/100, 11%), radiology (9/100, 9%), neurology, and neurosurgery (3/100, 3%). Other affiliations included the department of biomedical engineering (3/100, 3%), department of depression and mood disorder (3/100, 3%), department of electrical and computer engineering (3/100, 3%), college of mechatronics and automation (1/100, 1%), and school of mathematical sciences and center for computational systems biology (1/100, 1%).

Journals

The top 100 most highly cited articles were published in 28 international journals. The top three journals that published these articles were Biological Psychiatry (27/100, 27%), American Journal of Psychiatry (14/100, 14%), and Archives of General Psychiatry (9/100, 9%) and representing 50% of all journals. The journals and their recent (IFs) are shown in Table 3. The top three journals in terms of IF were Nature Medicine (53.44), Nature (49.96), and Jama Psychiatry (21.6).

Table 3: The Journals (2020) Impact Factor (IF) and Ranking

Discussion

The characteristics of neuroimaging in exploring the diagnosis, mechanism, and prognosis of depression were reported in this study using a bibliometric analysis of the 100 most frequently referenced research publications. The most highly cited articles (78 %) were in depression mechanism-related studies, followed by depression therapeutics outcome-related studies (13 %) using various neuroimaging techniques. Future research directions in this field could include investigating the pathophysiological mechanisms of depression using high-level evidence design and fMRI neuroimaging modalities (63 %). Our findings may be useful to new researchers and students in this field and to established scholars interested in trying new research methods. Meanwhile, our findings can help radiologists and neurologists understand the value of multidisciplinary collaboration between different disciplines in the clinical field of neuroradiology [15].

Since neuroimaging was first used in a depression study in 1992, this study assessed the bibliometric information of the top 25% of articles in order of yearly citation counts in brain imaging for depression [16]. According to our findings, there is still a lack of consensus on the quality of highly cited papers worldwide. According to the Levitt et al. analysis, the top 25% of annual citation counts were chosen as highly cited articles since they covered enough publications in the area while still satisfying the definition of highly cited papers [17]. As a result, we used the same criterion, namely the top 25%. Figure 1 demonstrates that no papers were included before 1992, even though no time constraint was specified for this investigation. Only one article was identified before 1997, with yearly citation counts ranging from 0.10 to 0.97, indicating that these articles did not qualify for the top 25% of the list.

According to one of the most cited papers, structural MRI was the first neuroimaging modality in depression that could find accountable therapeutic outcome.[18] Tissues can be distinguished in sMRI under an external magnetic field due to differences in signal from hydrogen proton. DTI was frequently used as a quantitative tool to study microstructural modification, pathophysiological process, diagnostic biomarkers, and assess the efficacy outcome in depression in the 21st century, with the rapid growth of radiological technology.[19] Nowadays, fMRI examinations provided functional information, metabolic changes were detected by PET and SPECT, and perfusion MR reflected vascular hemodynamic information about depression.[20,21] The variety of neuroimaging approaches and results emphasize the need for and value of innovative multivariable analytical methods that combine results from multiple modalities to improve diagnostic and mechanistic understanding accuracy, which could be a critical step in neuroimaging's full integration into the clinical realm of depression[22].

Among depression diagnosis investigations, earlier observation research demonstrated the diagnostic efficacy of imaging examination in detecting the biomarker for depressive patients treated with antidepressant medication according to the DSM-IV.[23] Meanwhile, research into image performance and characteristics appeared to aid in understanding the causes of depression and establishing a link between brain anomalies and clinical characteristics.[24] Studies primarily focused on the relevant relationship between structural alterations and other functions, including behavior and cognitive function, after introducing MRI quantitative technology.[25] The focus of recent studies has changed aside from etiology identification and toward methodically summarizing and detecting the relationship between brain abnormalities and potential etiology. [26] Furthermore, recent evidence suggests that depression cases are genetically linked [27]. In depression research, combining molecular genetic approaches with neuroimaging to discover specific brain abnormalities has become a new trend [28].

There are over 300 million people worldwide that suffer from depression.[29] Patients with early-onset depression are more likely to experience neurobehavioral disorders, and the functional brain abnormalities are particularly susceptible to negative outcomes. [30] However, due to a lack of effective means to track these changes, changes in the structure and function of depression remain poorly understood. These alterations have been clarified due to recent rapid advances in MRI technology. According to the most cited articles, fMRI was the first MR imaging method to locate the responsible focus in the pathophysiology of depression [31]. The focus of the study has recently switched from partial brain tissue volume measurements to a systematic examination of changes in total brain oxygen consumption and metabolic activity linked with neuropsychology using fMRI. The first study describing mechanism studies was published in 1997, according to the current review Neuroimaging in Depression [32]. Currently, functional information is obtained through fMRI tests, metabolic changes are evaluated through PET and SPECT findings, and vascular hemodynamic data are reflected through perfusion imaging (PI). Such numerous imaging techniques can thus provide information on depression's macro-structural alterations and functional and micromolecular information. Our findings show that the use of fMRI in pathophysiology and mechanism investigations is on the rise, with the increase being especially noticeable in recent years. Consequently, we believe that using fMRI to assess depression pathogenesis will be a major future research hotspot.

In the study of depression processes, fMRI is widely used. The majority of this research focused on drug-resistant depression, a topic that has sparked significant concern.[31] Many studies have investigated the functional connectivity network to learn more about the pathophysiological mechanisms of depression and, as a result, the causes of cognitive dysfunction. [33] It has been suggested that changes in default-mode function are linked to cognitive decline.[33] However, very few studies have taken advantage of fMRI modalities that can provide quantitative information on brain function.[15] The use of fMRI modalities to research the pathophysiology of depression may be on-trend. Our findings support a previous finding that neuroimaging research is a major focus in depression studies [13]. The top 100 most cited original articles had at least four decades of yearly citations, showing a high technical activity level. Recently, new imaging techniques have increased, as seen by historical trends in neuroimaging modalities. The diversity of neuroimaging tests utilized for PET imaging and the various disorder-related tasks evaluated in the fMRI studies demonstrate how adaptable various imaging modalities may be. These findings reflect the field's ongoing efforts to expand the area of research.

A study on pharmacotherapy was the top-ranked publication on the topic of depression. Sidney H et al. produced the top-ranked article on depression in 2001, a comparative study of major depressive disorder. At the time of the study, none of the patients had a concomitant DSM-IV diagnosis, and none was taking any extra psychiatric medication. Medical stability was also essential for the participants [21]. In depression, drug resistance is invariably accompanied by hippocampal neuronal loss, decreased dentate gyrus neurogenesis, and glial cell loss [34].

We also reported that the most cited articles on depression were published in Biological Psychiatry, which is perhaps related to the fact that the scope of the journal related to the pathophysiology and mechanism of the disease among the depression-specific articles belong to the American journal of psychiatry and archives of general psychiatry, which are of vital significance in shaping the landscape of depression mechanism, diagnosis, and management. Other high IF journals in the top 100 also provide several important journals, such as Nature Medicine, Nature, Jama Psychiatry, and the American Journal of Psychiatry. These prestigious journals tend to be cited more frequently in original articles and have had a high impact in recent years because the number of references allowed in most journals is limited.

Moreover, we found that about half of the 100 top-cited articles on depression originated from institutions in the United States of America, followed by the United Kingdom, Germany, China, Netherlands, Canada, Ireland, France, Australia, Italy, and Switzerland, Singapore, and Norway. Although most studies on depression using neuroimaging are from developed countries, the low-income and middle-income countries still have a similar prevalence of depression (59%) as that in high-income countries (55%), indicating that major depressive disorder is neither a simple consequence of modern-day lifestyle in developed countries nor a result of poverty [35].

The distribution of depressive patients among countries is almost equal, with an increasing trend in developing countries. Therefore, developing countries need a forward-thinking economy and a strong healthcare system. There are some limitations to this study. First, this bibliometric analysis was conducted using the Web of Science All Databases; the results may not apply to all depression articles found in other databases, such as Google Scholar. Second, the amount of citations is heavily influenced by the passage of time. Third, the rate of self-citation may have an impact on the outcomes. However, the number of yearly citations for the top 100 most referenced publications was limited; therefore, the results were unaffected.

Conclusion

In conclusion, this study provided useful information for future research collaborations between researchers and institutions and identified hot topics and trends in depression research. Although neuroimaging appears to be useful in understanding the pathophysiology of depression, additional research is needed. Hence, further research is needed, including high-quality randomized controlled trials that follow defined guidelines and have a minimal risk of bias.

Conflict of Interest

The authors declare that no commercial or financial relationships could be construed as a potential conflict of interest during the research.

Fund

This work was not supported by any fund.

References

1. Hofmann S G, Curtiss J, Carpenter J K, et al. Effect of treatments for depression on quality of life: a meta-analysis. Cogn Behav Ther. 46 (2017): 265-286.
2. WHO, “Depression.” https://www.who.int/news-room/fact-sheets/detail/depression
3. Zenebe Y, Akele B, Selassie M W, et al. Prevalence and determinants of depression among old age: a systematic review and meta - analysis. Ann Gen Psychiatry (2021): 1-19.
4. Kuo D C, Tran M, Shah A A, et al. Depression and the Suicidal Patient. Emerg Med Clin North Am, (2015): 765-778.
5. Li L, Wu C, Gan Y, et al. Insomnia and the risk of depression: A meta-analysis of prospective cohort studies. BMC Psychiatry (2016).
6. Guo Y, Sun J, Hu S, et al. Hospitalization costs and financial burden on families with children with depression: A cross-section study in Shandong Province, China. Int J Environ Res. Public Health (2019).
7. Blakeman K, .Bibliometrics in a digital age: help or hindrance. Sci Prog (2018): 293-310.
8. Ahmad P, Dummer P M H, Noorani T Y et al. The top 50 most-cited articles published in the International Endodontic Journal. Int Endod. (2019): 803-818.
9. Wu Y, Jin X, Xue Y, et al. Evaluation of research topic evolution in psychiatry using co-word analysis. Med. (United States), 25 (2017).
10. Wang, C Y. The top-100 highly cited original articles on drug therapy for ventilator-associated pneumonia,” Front. Pharmacol, 10 (2019):1-7.
11. Lin C H, Chien T W, Yan, Y H. Predicting the number of article citations in the field of attention-defcit/hyperactivity disorder (ADHD) with the 100 top-cited articles since 2014: a bibliometric analysis,” Ann. Gen. Psychiatry, 20 (2014):1-7.
12. Popkirov S, Jungilligens J, Schlegel U, et al. Research on dissociative seizures: A bibliometric analysis and visualization of the scientific landscape,” Epilepsy Behav, 83, (2018):162-167.
13. Gong, B. Neuroimaging in Psychiatric Disorders?: A Bibliometric Analysis of the 100 Most Highly Cited Articles,” J Neuroimaging, (2019).
14. Brandt J S, Downing A C, Howard D L, et al. Citation classics in obstetrics and gynecology?: the 100 most frequently cited journal articles in the last 50 years,” YMOB, 203 (2010): 355-355.
15. Li J, Duan X, Cui Q, et al. More than just statics: Temporal dynamics of intrinsic brain activity predicts the suicidal ideation in depressed patients,” Psychol. Med, 49 (2019): 852-860.
16. Drevets W C, Videen T O, Price J L, et al. A functional anatomical study of unipolar depression,” J. Neurosci. 12 (1992) 3628-3641.
17. Levitt J M, and Thelwall, M. Does the higher citation of collaborative research differ from region to region? A case study of Economics,” Scientometrics, (2010):171-183.
18. Pande A C, Grunhaus L J, Aisen A M, et al. A preliminary magnetic resonance imaging study of ECT-treated depressed patients,” Biol. Psychiatry, 27 (1990): 102–104.
19. Versace A. Elevated left and reduced right orbitomedial prefrontal fractional anisotropy in adults with bipolar disorder revealed by tract-based spatial statistics,” Arch. Gen. Psychiatry, 65 (2008): 1041-1052.
20. Malison R T. Reduced brain serotonin transporter availability in major depression as measured by 2β-carbomethoxy-3β-(4-iodophenyl tropane and single photon emission computed tomography,” Biol. Psychiatry, 44 (1998): 1090-1098.
21. Kennedy S H. Changes in regional brain glucose metabolism measured with positron emission tomography after paroxetine treatment of major depression,” Am. J. Psychiatry, 158 (2001): 899-905.
22. Linden D E J. The Challenges and Promise of Neuroimaging in Psychiatry,” Neuron, 73 (2011): 8-22.
23. Bremner J D, Narayan M, Anderson E R L, et al. Hippocampal volume reduction in major depression,” Am. J. Psychiatry, 157 (2000): 115–117.
24. Shi Y. A Study of the Brain Abnormalities of Post-Stroke Depression in Frontal Lobe Lesion,” Sci. Rep. 7 (2017): 1-10.
25. Zhang F F, Peng W, Sweeney J A, et al. Brain structure alterations in depression: Psychoradiological evidence,” CNS Neurosci. Ther., 24 (2018): 994–1003.
26. Peng W. Brain structural abnormalities in emotional regulation and sensory processing regions associated with anxious depression,” Prog. Neuro-Psychopharmacology Biol. Psychiatry, 94 (2019): 109676.
27. Peña C J, Bagot R C, Labonté B, et al. Epigenetic signaling in psychiatric disorders,” J. Mol. Biol, 426 (2014): 3389–3412.
28. Shelton R C. The Molecular Neurobiology of Depression,” Psychiatr. Clin. North Am., 30 (2007): 1-11.
29. Hamel C. Screening for depression in women during pregnancy or the first year postpartum and in the general adult population?: a protocol for two systematic reviews to update a guideline of the Canadian Task Force on Preventive Health Care,” Psychol. Med, (2019): 1-13.
30. Belden A C. Anterior insula volume and guilt: Neurobehavioral markers of recurrence after early childhood major depressive disorder,” JAMA Psychiatry, 72 (2015): 40-48.
31. Downar J Anhedonia and reward-circuit connectivity distinguish nonresponders from responders to dorsomedial prefrontal repetitive transcranial magnetic stimulation in major depression,” Biol. Psychiatry, 76 (2013): 176-185.
32. Krishnan K R R, Hays J C, and Blazer D G. MRI-defined vascular depression,” Am J Psychiatry, 6 (1997): 115-119.
33. Zhu X. Evidence of a dissociation pattern in resting-state default mode network connectivity in first-episode, treatment-naive major depression patients,” Biol. Psychiatry, 71 (2012): 611-617.
34. Arnone D. State-dependent changes in hippocampal grey matter in depression,” Mol. Psychiatry, 18 (2013): 1265–1272.
35. Malhi G S and Mann J J. Seminar Depression,” Lancet, 6736 (2018).