Over time for the past few years, facial expression identification has been a promising area. However, darkness, lighting conditions, and other factors make facial emotion identification challenging to detect. As a result, thermal images are suggested as a solution to such problems and for a variety of other benefits. Furthermore, focusing on significant regions of a face rather than the entire face is sufficient for reducing processing and improving accuracy at the same time. This research introduces novel infrared thermal image-based approaches for facial emotion recognition. First, the entire image of the face is separated into four pieces. Then, we accepted only four active regions (ARs) to prepare training and testing datasets. These four ARs are the left eye, right eye, and lips areas. In addition, ten-folded cross-validation is proposed to improve recognition accuracy using Convolutional Neural Network (CNN), a machine learning technique. Furthermore, we incorporated a parallelism technique to reduce processing-time in testing and training datasets. As a result, we have seen that the processing time reduces to 50%. Finally, a decision-level fusion is applied to improve the recognition accuracy. As a result, the proposed technique achieves a recognition accuracy of 96.87 %. The achieved accuracy ascertains the robustness of our proposed scheme.
Citation: Basem Assiri, Mohammad Alamgir Hossain. Face emotion recognition based on infrared thermal imagery by applying machine learning and parallelism[J]. Mathematical Biosciences and Engineering, 2023, 20(1): 913-929. doi: 10.3934/mbe.2023042
[1] | Kholoud Baraka, Rania Abozahra, Maged Wasfy Helmy, Nada Salah El Dine El Meniawy, Sarah M Abdelhamid . Investigation of the protective and therapeutic effects of Lactobacillus casei and Saccharomyces cerevisiae in a breast cancer mouse model. AIMS Microbiology, 2022, 8(2): 193-207. doi: 10.3934/microbiol.2022016 |
[2] | Md. Aoulad Hosen, Fozol Korim Ovi, Harunur Rashid, MD. Hasibul Hasan, Md. Abdul Khalek, Mahmudul Hasan, Farhana Easmin, Nazmi Ara Rumi, Mohammad Shariful Islam . Characterization of Vibrio spp. in environmental water samples collected from flood prone areas of Bangladesh and their antibiotic resistance profile. AIMS Microbiology, 2021, 7(4): 471-480. doi: 10.3934/microbiol.2021028 |
[3] | Yue Tang, Shaun Cawthraw, Mary C. Bagnall, Adriana J. Gielbert, Martin J. Woodward, Liljana Petrovska . Identification of temperature regulated factors of Campylobacter jejuni and their potential roles in virulence. AIMS Microbiology, 2017, 3(4): 885-898. doi: 10.3934/microbiol.2017.4.885 |
[4] | Tatyana V. Polyudova, Daria V. Eroshenko, Vladimir P. Korobov . Plasma, serum, albumin, and divalent metal ions inhibit the adhesion and the biofilm formation of Cutibacterium (Propionibacterium) acnes. AIMS Microbiology, 2018, 4(1): 165-172. doi: 10.3934/microbiol.2018.1.165 |
[5] | Ashok Chakraborty, Anil Diwan, Jayant Tatake . Prospect of nanomaterials as antimicrobial and antiviral regimen. AIMS Microbiology, 2023, 9(3): 444-466. doi: 10.3934/microbiol.2023024 |
[6] | Arsenio M. Fialho, Nuno Bernardes, Ananda M Chakrabarty . Exploring the anticancer potential of the bacterial protein azurin. AIMS Microbiology, 2016, 2(3): 292-303. doi: 10.3934/microbiol.2016.3.292 |
[7] | Nieves Barros, Sergio Feijoo, César Pérez-Cruzado, Lee D. Hansen . Effect of soil storage at 4 °C on the calorespirometric measurements of soil microbial metabolism. AIMS Microbiology, 2017, 3(4): 762-773. doi: 10.3934/microbiol.2017.4.762 |
[8] | Paola Ortiz-Ojeda, Katty Ogata-Gutiérrez, Doris Zúñiga-Dávila . Evaluation of plant growth promoting activity and heavy metal tolerance of psychrotrophic bacteria associated with maca (Lepidium meyenii Walp.) rhizosphere. AIMS Microbiology, 2017, 3(2): 279-292. doi: 10.3934/microbiol.2017.2.279 |
[9] | Laurent Coquet, Antoine Obry, Nabil Borghol, Julie Hardouin, Laurence Mora, Ali Othmane, Thierry Jouenne . Impact of chlorhexidine digluconate and temperature on curli production in Escherichia coli—consequence on its adhesion ability. AIMS Microbiology, 2017, 3(4): 915-937. doi: 10.3934/microbiol.2017.4.915 |
[10] | MP Lizardo, FK Tavaria . Probiotic growth in skin-like conditions. AIMS Microbiology, 2022, 8(4): 388-402. doi: 10.3934/microbiol.2022027 |
Over time for the past few years, facial expression identification has been a promising area. However, darkness, lighting conditions, and other factors make facial emotion identification challenging to detect. As a result, thermal images are suggested as a solution to such problems and for a variety of other benefits. Furthermore, focusing on significant regions of a face rather than the entire face is sufficient for reducing processing and improving accuracy at the same time. This research introduces novel infrared thermal image-based approaches for facial emotion recognition. First, the entire image of the face is separated into four pieces. Then, we accepted only four active regions (ARs) to prepare training and testing datasets. These four ARs are the left eye, right eye, and lips areas. In addition, ten-folded cross-validation is proposed to improve recognition accuracy using Convolutional Neural Network (CNN), a machine learning technique. Furthermore, we incorporated a parallelism technique to reduce processing-time in testing and training datasets. As a result, we have seen that the processing time reduces to 50%. Finally, a decision-level fusion is applied to improve the recognition accuracy. As a result, the proposed technique achieves a recognition accuracy of 96.87 %. The achieved accuracy ascertains the robustness of our proposed scheme.
Parkinson’s disease (PD) is an incurable, progressive movement disorder affecting approximately one million Americans (www.pdf.org) and is the 14th leading cause of death in the US (www.cdc.gov). The pathogenesis of PD involves the degeneration of neurons, especially dopaminergic neurons in the substantia nigra of the midbrain [1], and the presence of Lewy bodies/neuritis in various brain regions [2]. Deficiency of the nigro-striatal pathways can cause dopamine depletion-specific symptoms such as motor dysfunctions and multiple non-motor clinical issues. The clinical diagnosis of PD is based on the four cardinal symptoms: tremor, rigidity, bradykinesia, and postural instability [3], which are delayed and subtle. Since the key pathology of PD occurs in the brainstem which is difficult to assess in living patients, it is challenging to monitor PD progression in an unbiased and objective manner. Thus, developing reliable clinical diagnostic methods that are feasible, objective, measurable, specific and sensitive for PD is essential to improve the clinical management of the disease.
A molecular biomarker is a biological molecule found in blood, other bodily fluids or tissues that can be measured as an indicator of a normal or abnormal biological process, or of a disease condition. Molecular biomarkers can be used as a diagnostic or prognostic tool for a specific disease. In PD, biochemical biomarkers which are heavily based on alpha-synuclein (SNCA), the primary structural component of Lewy bodies, have been proposed in the past few years [4]. This mini review will focus on the most recent development of non-SNCA, microRNA (miRNA)-based biomarkers identified in various biofluids for PD diagnosis.
miRNAs are a family of short (18-22 nucleotides), single-stranded non-coding RNAs that act as gene expression modulators at the post-transcriptional level to fine-tune the expression of specific messenger RNAs (mRNAs) [5, 6]. Many miRNAs can target multiple mRNAs by recognizing short, specific complementary sequences in the mRNAs and control the expression of hundreds to thousands of genes [7, 8]. miRNA expression profiling has been used frequently to identify biomarkers and/or molecular pathways associated with specific human diseases. miRNAs are also known to be highly abundant, tissue-specific, quantifiable, highly stable with non-post process modification, which are “ideal” characteristics of biomarkers. In neurodegeneration, although the pathogenic mechanisms are complex and not fully defined, deregulation of miRNAs has been increasingly recognized to play important roles in the process [9]. There is also emerging evidence describing the important roles of miRNAs in neuronal processes such as neuronal differentiation and neuroplasticity. miRNAs as potential clinical biomarkers for PD has been described extensively in our previous review [10].
Biological fluids or biofluids are liquids originating from human bodies. They are an excellent source for biomarker research as their close proximity to cells can potentially reflect their biochemical contents and hence the biological conditions of the cells [11]. Common biofluids used in biomarker research include peripheral blood, blood serum or plasma, saliva, urine, and cerebral spinal fluid (CSF). Deregulation of blood miRNAs has been significantly correlated with disease-associated genetic variants [12]. Since biofluid sample collection is inexpensive, rapid, simple, and minimally invasive (except for CSF), it has a large advantage over other solid or semi-solid biospecimens for studying disease biomarkers.
Blood is an ideal source for biomarkers because it can be obtained in a simple, fast, low-cost and minimally invasive manner. The first report of peripheral blood-based miRNA biomarkers for PD was demonstrated in 2011 by a quantitative real-time PCR (qRT-PCR) approach [13]. A set of 3 miRNAs (miR-1, miR-22*, and miR-29a) with lower miRNA expression was identified when compared between de novo and treated PD vs. control and early-onset PD. A second set of 3 miRNAs (miR-16-2*, miR-26a2*, and miR-30a) showed 50% higher expression in treated PD vs. control subjects (Table 1). The second report applied a microarray approach to identify 18 under-expressed miRNAs in peripheral blood mononuclear cells of PD vs. control [14]. The third report used a combination of small RNA next-generation sequencing (small RNA-seq) and exon microarray technologies to investigate miRNA expression in blood leukocytes (white blood cells) of healthy controls as well as PD patients before and after deep brain stimulation (DBS) treatment [15]. A set of 5 miRNAs (miR-4293, miR-378c, miR-18b*, miR-20a, and miR-1249) that are under-expressed in PD vs. DBS has inversed/higher expression in controls whe n compared with PD, suggesting both treatment and disease related changes in miRNA expression. Interestingly, down-regulation of miR-29a, miR-29c, miR-30c, and miR-19b can also be observed in white blood cells of PD DBS-off vs. DBs-on patients and/or in the serum of PD vs. control (see Table 1 and cmiRNAs in plasma/serum section below). Unfortunately, detection sensitivity of miRNA expression among biofluid types (i.e., peripheral blood vs. mononuclear cells vs. white blood cells vs. serum) could not be evaluated mainly due to different technologies (i.e.,qRT-PCT, microarray and next-generation sequencing) used among these studies.
miRNA | Biofluid type | Sample size | Sample type | Method | Reference | |
↓ miR-29a | Peripheral blood White blood cells Serum | 15 PD vs. 8 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | qRT-PCR (SYBR green) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Margis et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-29c | Peripheral blood mononuclear cell White blood cell Serum | 19 PD vs.13 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-30c | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Vallelunga et al., 2014 | |
↓ miR-19b | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Botta-Orfila et al., 2014 | |
↓ miR-1, miR-22* ↑ miR-16-2*, miR-26a2*, miR-30a | Peripheral blood | 15 PD vs. 8 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD levodopa/carbidopa treated vs. CTR | qRT-PCR (SYBR green) | Margis et al., 2011 | |
↓ miR-335, miR-374a, miR-199a-3p/miR-199b-3p, miR-126*, miR-151-3p, miR-199a-5p, miR-151-5p, miR-126, miR-29b, miR-147, miR-28-5p, miR-30b, miR-374b, miR-301a, miR-26a | Peripheral blood mononuclear cell | 19 PD vs.13 CTR | PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) | Martins et al., 2011 | |
↓ miR-320a/b/c, miR-769, miR-92b, miR-16 ↑ miR-199b, miR-1274b, miR-21, miR-150, miR-671, miR-1249, miR-20a, miR-18b*, miR-378c, miR-4293 ↓ miR-4293, miR-378c, miR-18b*, miR-424*, miR-20a, miR-210, miR-1249, miR-93 ↑ miR-4317, miR-143, miR-424 ↓ miR-652, miR-15a*, miR-376c, miR-143 ↑ miR-423, miR-365, miR-486, miR-1260, miR-218 | White blood cell | 7 PD vs. 6 CTR | PD (pre-DBS) vs. CTR PD (pre-DBS) vs. PD (post-DBS) PD (DBS-off) vs PD (DBS-on) | Small RNA-sequencing (ABI SOLiD) | Soreq et al., 2013 | |
A panel of four miRNA biomarkers ↑ miR-1826, ↓ miR-450b-3p ↑ miR-626, miR-505 | Plasma | Discovery set: 32 PD vs.32 CTR Replication set: 42 PD vs.30 CTR Validation set: 30 PD, 4 MSA, 5 PSP vs. 8 CTR | PD (treated) vs. CTR | Agilent miRNA microarray v.3, validation by qRT-PCR (TaqMan) | Khoo et al., 2012 | |
↑ miR-331-5p | Plasma | 31 PD vs. 25 CTR | PD (non-treated) vs. CTR | TaqMan Low Density miRNA array (3 pools of PD and CTR), then qRT-PCR (TaqMan) | Cardo et al., 2013 | |
↓ miR-339-5p ↑ miR-223*, miR-324-3p, miR-24 ↓ miR-148b ↑ miR-223*, miR-324-3p, miR-24 | Serum | Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD + MSA vs. CTR PD vs. CTR | TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Vallelunga et al., 2014 | |
↓ miR-19a | Serum | Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR | TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Botta-Orfila et al., 2014 | |
↓ miR-16-2-3p, miR-1294 ↑ miR-338-3p, miR-30e-3p, miR-30a-3p ↓ miR-132-5p, miR-485-5p, miR-127-3p, miR-128, miR-409-3p, miR-433, miR-370, miR-431-3p, miR-873-3p, miR-136-3p, miR-212-3p, miR-10a-5p, miR-1224-5p, miR-4448 ↑ miR-19a-3p, miR-19b-3p, let-7g-3p | Serum (post mortem) CSF (post mortem) | 50 PD vs. 62 CTR 57 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Small RNA Sequencing (Illumina HiSeq2000) | Burgos et al., 2014 | |
PD: Parkinson’s disease; CTR: healthy control; PSP: progressive supranuclear palsy; MSA: multiple system atrophy; LRRK2 PD: G2019S mutation in PD; DBS: Deep brain stimulation; ↓: down-regulated miRNA expression; ↑: up-regulated miRNA expression. |
The presence of circulating miRNAs (cmiRNAs) was first described in the serum of diffuse large B-cell lymphoma patients [16]. Since then, many studies have shown that miRNAs exist not only at the cellular level but also as circulating, cell-free miRNAs in serum, plasma, and CSF. There are several hypotheses of how cmiRNAs can be released into circulatory systems: 1) direct leakage of cellular miRNAs as by-products of tissue damage or apoptosis due to certain pathological conditions (such as metastasis or inflammation), 2) active secretion by cells as free miRNAs with or without cell stimulation, and 3) secretion in the form of cell-derived microvesicles (eg. microparticles and exosomes), apoptotic bodies, miRNA-protected proteins (eg. Ago2 and NPM1) or high-density lipoprotein [17, 18, 19, 20, 21, 22].
Similar to miRNAs, cmiRNAs are highly stable and resistant to RNase activity, extreme pH and temperature [23, 24]. cmiRNA expression are also known to correlate with various human diseases, especially cancers [17]. In PD, we performed a proof-of-concept study and first reported a panel of four cmiRNA as potential PD biomarkers [25]. With high stability and capacity to reflect disease status, cmiRNAs have great potential as molecular biomarkers for diagnosis and prognosis of neurodegenerative diseases. Readers can further refer to a review by Sheinerman and Umansky [26] on cmiRNAs biomarkers for various neurodegenerative diseases and three common approaches used for such studies. In Alzheimer’s disease (AD), miRNA expression changes in brain and CSF have provided insights into the disease pathways [27]. Therefore, besides their utility as disease biomarkers, cmiRNA may provide new insights for the pathogenesis of PD.
We published the first report of four plasma-based cmiRNAs (miR-1826, miR-450b-3p, miR-626, and miR-505) as a panel of potential biomarkers for PD using Agilent microarrays followed by a qRT-PCR approach [25]. A validation study using a new, independent cohort has been carried out to confirm the reproducibility and robustness of these candidate biomarkers. One year later, another group used the TaqMan low-density microarray cards and qRT-PCR and found miR-331-5 significantly up-regulated in plasma of PD patients compared with healthy controls [28]. Both findings are mutually exclusive and could be explained by used of 1) different discovery platforms which may have different detection sensitivities, 2) different statistical algorithms to identify differentially-expressed cmiRNAs, and 3) treated PD patients in the formal and naïve PD patients in the later report.
Recently, three additional studies on serum-based cmiRNA biomarkers for PD have been reported. Four differentially-expressed cmiRNAs (miR-399-5p, miR-223*, miR-324-3p, and miR-24) were identified and validated in PD and MSA (multiple system atrophy) vs. control subjects [29]. The same cmiRNAs (except miR-339-5p) were identified in PD vs. control, in addition to miR-30c and miR-148b which were down-regulated. Interestingly, miR-30c was also found down-regulated in the peripheral blood mononuclear cell of PD when compared with control. Another report used PD and LRRK2 (leucine-rich repeat kinase 2) PD as discovery and first validation cohort to identify 4 statistically significant cmiRNAs (miR-29a, miR-29c, miR-19a, and miR-19b) in PD vs. control in a second validation study [30]. Strikingly, down-regulation of miR-29a, miR-29c, miR-30c, and miR-19b were also shown in previous blood-based biomarker studies [13, 14]. Repeated detection of similar miRNAs regardless of biofluid types and detection technologies make these miRNAs strong biofluid-based biomarker candidates for PD which need to be validated. In another miRNA expression profiling study using small next-generation sequencing on post mortem sera, 5 differentially-expressed cmiRNAs were identified between PD and control [31].
Small RNA-seq was also recently used to detect 17 differentially-expressed cmiRNAs in post mortem CSF of PD or AD patients vs. healthy controls [31]. Although one differentially-expressed cmiRNA was identified in both CSF and serum for AD, there was no overlapping cmiRNA candidate in CSF and serum from the same PD patients. This finding suggests that cmiRNA expression in serum may not be corresponding to those in CSF which is directly connected to the central nervous system; CSF and serum may not provide similar cmiRNA expression profiles in this case. Moreover, the viability of cmiRNA expression in postmortem vs. fresh serum and CSF is relatively unknown and worth investigation.
Table 1 summarizes the biofluid-based miRNA studies for PD mentioned above.
Biofluid-based miRNA biomarker research for PD has been emerging rapidly. cmiRNAs are envisioned to become potential biomarkers for early detection and progression of PD. However, there are several impeding factors that may come into play. First, miRNAs usually exist in much lower amounts in biofluids than tissues. Hence, depending on the sensitivity of the detection techniques, they can remain undetected and be missed as potential biomarkers. Fortunately, recent advancements of non-biased amplification and enrichment techniques have substantially improved the detection sensitivity of miRNA expression. Second, biomarker research is known for its complexity and difficulty in correlating different biomarker data from different reports. As shown in Table 1, biomarkers in bodily fluids are screened and analyzed by a variety of techniques. Variations may also involve sample collection and storage, sample preparation, data normalization and data analysis [10]. Therefore, a unifying method and/or technique should be applied for each specific biomarker in general. However, in a “real-world” clinical setting, standardization from the beginning of sample collection to assay processes may not be feasible or realistic. This also brings us to the third limitation where either small or no validation studies were performed to confirm the reproducibility and robustness of these biofluid-based biomarker candidates and its assays; this drawback has further delayed the incorporation of molecular biomarkers into routine preclinical or clinical applications. Fourth, the presence of miRNAs in biological fluids as a disease indicator remains a subject of considerable debate. Although deregulation of miRNAs has been increasingly recognized to play important roles in neurodegeneration [9], the potential of utilizing miRNAs to reflect pathogenic changes in the brain needs to be further investigated and validated.
This mini review shows that miRNAs in bodily fluids hold promise as potential molecular biomarkers for PD. Unique characteristics of miRNAs and cmiRNAs make them highly attractive as molecular biomarkers. Follow-up validation studies and standardization are critical to ensure successful development of sensitive and specific biomarkers for PD. The finding of stable cmiRNAs in biofluids suggests their possible function as mediators in cell-cell communication in response to biological stimuli [32]. Thus, it is possible that cmiRNAs in serum or plasma may directly or indirectly reflect the pathological status of the brain. It is also known that miRNAs can be transported by exosomes which are lipid microvesicles of endocytic origin secreted by many cell types, including neurons, astrocytes, oligodendrocytes, and microglia [33, 34, 35, 36]. Recent evidence has shown the involvement of exosomes in the pathogenesis of PD [37]. Although the association of exosome-based miRNAs in neurodegeneration has not been established, it is worth future investigation.
In AD, cmiRNAs in CSF and serum have been identified to correlate with the severity of disease such as Braak stages, neurofibrillary tangle and plaque-density scores [27, 31]. Hence, future investigation of cmiRNA expression in correlation with PD severity (e.g. SNCA accumulation or Lewy bodies) may accelerate discovery of progression biomarkers for PD. In summary, establishing and validating biofluid-based miRNA biomarkers may facilitate the development of objective diagnostic and prognostic molecular tools for PD to translate basic research discovery to clinical setting.
S.K. Khoo is supported by grants from the Michael J. Fox Foundation for Parkinson’s Research.
All authors declare no conflict of interest.
[1] | M. A. Hossain, B. Assiri, Facial emotion verification by infrared image, in International Conference on Emerging Smart Computing and Informatics (ESCI), (2020). https://doi.org/10.1109/ESCI48226.2020.9167616 |
[2] | M. A. Hossain, B. Assiri, Emotion specific human face authentication based on infrared thermal image, in 2020 2nd International Conference on Computer and Information Sciences (ICCIS), (2020). https://doi.org/10.1109/ICCIS49240.2020.9257683 |
[3] | M. Vollmer, K. P. Möllmann, Infrared Thermal Imaging: Fundamentals, Research and Applications, 2nd edition, WILEY-VCH Verlag, 2018. https://doi.org/10.1002/9783527693306 |
[4] |
M. A. Hossain, G. Sanyal, Tracking humans based on interest point over span-space in multifarious situations, Int. J. Software Eng. Appl., 10 (2016), 175–192. https://doi.org/10.14257/ijseia.2016.10.9.15 doi: 10.14257/ijseia.2016.10.9.15
![]() |
[5] | C. Myeon-gyun, A study on the obstacle recognition for autonomous driving RC car using lidar and infrared thermal camera, in 2019 Eleventh International Conference on Ubiquitous and Future Networks (ICUFN), (2019). http://doi.org/10.1109/ICUFN.2019.8806152 |
[6] | Q. Wan, S. P. Rao, A. Kaszowska, V. Voronin, K. Panetta, H. A, Taylor, et al., Face description using anisotropic gradient: infrared thermal to visible face recognition, in Mobile Multimedia Image Processing, Security, and Applications, (2018). https://doi.org/10.1117/12.2304898 |
[7] |
T. Bae, K. Youngchoon, A. Sangho, IR-band conversion of target and background using surface temperature estimation and error compensation for military IR sensor simulation, Sensors, 19 (2019), 2455. https://doi.org/10.3390/s19112455 doi: 10.3390/s19112455
![]() |
[8] | Y. Abdelrahman, P. Knierim, P. W. Wozniak, N. Henze, A. Schmidt, See through the fire: evaluating the augmentation of visual perception of firefighters using depth and thermal cameras, in Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing, (2017), 693–696. https://doi.org/10.1145/3123024.3129269 |
[9] | M. A. Hossain, D. Samanta, G. Sanyal, Eye diseases detection based on covariance, Int. J. Comput. Sci. Inf. Technol. Sec., 2 (2012), 376–379. |
[10] |
E. Sousa, R. Vardasca, S. Teixeira, A. Seixas, J. Mendes, A. Costa-Ferreira, A review on the application of medical infrared thermal imaging in hands, Infrared Phy. Technol., 85 (2017), 315–323. https://doi.org/10.1016/j.infrared.2017.07.020 doi: 10.1016/j.infrared.2017.07.020
![]() |
[11] |
M. A. Hossain, B. Assiri, Facial expression recognition based on active region of interest using deep learning and parallelism, Peer. J. Comput. Sci., 8 (2022), e894. https://doi.org/10.7717/peerj-cs.894 doi: 10.7717/peerj-cs.894
![]() |
[12] |
N. M. Moacdieh, N. Sarter, The effects of data density, display organization, and stress on search prformance: An eye tracking study of clutter, IEEE Trans. Hmman Mach. Syst., 47 (2017), 886–895. https://doi.org/10.1109/THMS.2017.2717899 doi: 10.1109/THMS.2017.2717899
![]() |
[13] |
M. A. Hossain, B. Assiri, An enhanced eye tracking approach using pipeline computation, Arab. J. Sci. Eng., 45 (2020), 1–14. https://doi.org/10.1007/s13369-019-04322-7 doi: 10.1007/s13369-019-04073-5
![]() |
[14] |
S. U. Mahmood, F. Crimbly, S. Khan, E. Choudry, S. Mehwish, Strategies for rational use of personal protective equipment (PPE) among healthcare providers during the COVID-19 crisis, Cureus, 12 (2020), e8248. http://doi.org/10.7759/cureus.8248 doi: 10.7759/cureus.8248
![]() |
[15] |
C. Filippini, D. Perpetuini, D. Cardone, A. M. Chiarelli, A. Merla, Thermal infrared imaging-based affective computing and its application to facilitate human robot interaction: A review, Appl. Sci., 10 (2020), 2924. https://doi.org/10.3390/app10082924 doi: 10.3390/app10082924
![]() |
[16] |
M. A. Eid, N. Giakoumidis, A. El Saddik, A novel eye-gaze-controlled wheelchair system for navigating unknown environments: Case study with a person with ALS, IEEE Access, 4 (2016), 558–573. https://doi.org/10.1109/ACCESS.2016.2520093 doi: 10.1109/ACCESS.2016.2520093
![]() |
[17] | M. A. Hossain, D. Samanta, G. Sanyal, Extraction of panic expression depending on lip detection, in 2012 International Conference on Computing Sciences, (2012), 137–141, https://doi.org/10.1109/ICCS.2012.35 |
[18] |
M. A. Hossain, D. Samanta, Automated smiley face extraction based on genetic algorithm, Comput. Sci. Inf. Technol., 2012 (2012), 31–37. https://doi.org/10.5121/csit.2012.2304 doi: 10.5121/csit.2012.2304
![]() |
[19] |
S. S. Alam, R. Jianu, Analyzing eye-tracking information in visualization and data space: From where on the screen to what on the screen, IEEE Trans. Visualization Comput. Graphics, 23 (2017), 1492–1505. https://doi.org/10.1109/TVCG.2016.2535340 doi: 10.1109/TVCG.2016.2535340
![]() |
[20] |
W. Zhang, H. Liu, Toward a reliable collection of eye-tracking data for image quality research: Challenges, solutions, and applications, IEEE Trans. Image Process., 26 (2017), 2424–2437. https://doi.org/10.1109/TIP.2017.2681424 doi: 10.1109/TIP.2017.2681424
![]() |
[21] |
A. Torabi, G. Massé, G. A. Bilodeau, An iterative integrated framework for thermal–visible image registration, sensor fusion, and people tracking for video surveillance applications, Comput. Vision Image Understanding, 116 (2012), 210–221. https://doi.org/10.1016/j.cviu.2011.10.006 doi: 10.1016/j.cviu.2011.10.006
![]() |
[22] | Y. Liu, Y. Cao, Y. Li, M. Liu, R. Song, Y. Wang, et al., Facial expression recognition with PCA and LBP features extracting from active facial patches, in 2016 IEEE International Conference on Real-time Computing and Robotics (RCAR), (2016). https://doi.org/10.1109/RCAR.2016.7784056 |
[23] |
W. R. Almeida, F. A. Andaló, R. Padilha, G. Bertocco, W. Dias, R. da S. Torres, et al., Detecting face presentation attacks in mobile devices with a patch-based CNN and a sensor aware loss function, Plos One, 4 (2020), 1–24. https://doi.org/doi.org/10.1155/2020/6385281 doi: 10.1155/2020/6385281
![]() |
[24] | F. Khan, Facial expression recognition using facial landmark detection and feature extraction via neural networks, preprint, arXiv: 1812.04510. |
[25] | M. A. Hossain, H. Zogan, Emotion tracking and grading based on sophisticated statistical approach, Int. J. Adv. Electron. Comput. Sci., 5 (2018), 9–13. https://doi.org/12-451-152482928314-18 |
[26] |
W. Zhang, X. Sui, G. Gu, Q. Chen, H. Cao, Infrared thermal imaging super-resolution via multiscale Spatio-Temporal feature fusion network, IEEE Sensors J., 21 (2021), 19176–19185. https://doi.org/10.1109/JSEN.2021.3090021 doi: 10.1109/JSEN.2021.3090021
![]() |
[27] | H. Mady, S. M. S. Hilles, Face recognition and detection using Random forest and combination of LBP and HOG features, in 2018 International Conference on Smart Computing and Electronic Enterprise (ICSCEE), (2018). https://doi.org/10.1109/ICSCEE.2018.8538377 |
[28] | K. T. Islam, R. G. Raj, A. Al-Murad, Performance of SVM, CNN, and ANN with BoW, HOG, and image pixels in face recognition, in 2017 2nd International Conference on Electrical & Electronic Engineering (ICEEE), (2017). https://doi.org/10.1109/CEEE.2017.8412925 |
[29] |
M. Sajjad, S. Zahir, A. Ullah, Z. Akhtar, K. Muhammad, Human behavior understanding in big multimedia data using CNN based facial expression recognition, Mobile Networks Appl., 25 (2020), 1611–1621. https://doi.org/10.1007/s11036-019-01366-9 doi: 10.1007/s11036-019-01366-9
![]() |
[30] | P. Liu, S. Han, Z. Meng, Y. Tong, Facial expression recognition via a boosted deep belief network, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2014). https://doi.org/10.1109/CVPR.2014.233 |
[31] |
A. T. Lopes, E. Aguiar, A. F. D. Souza, T. O. Santos, Facial expression recognition with convolutional neural networks: coping with few data and the training sample order, Pattern Recognit., 61 (2017), 610–628. https://doi.org/10.1016/j.patcog.2016.07.026 doi: 10.1016/j.patcog.2016.07.026
![]() |
[32] |
S. Rajan, P. Chenniappan, S. Devaraj, N. Madian, Novel deep learning model for facial expression recognition based on maximum boosted CNN and LSTM, IET Image Process., 14 (2020), 1373–1381. https://doi.org/10.1049/iet-ipr.2019.1188 doi: 10.1049/iet-ipr.2019.1188
![]() |
[33] |
M. A. Hossain, G. Sanyal, A new improved tactic to extract facial expression based on genetic algorithm and WVDF, Int. J. Adv. Inf. Technol., 2 (2012), 37. https://doi.org/10.5121/ijait.2012.2504 doi: 10.5121/ijait.2012.2504
![]() |
[34] | M. A. Hossain, D. Samanta, G. Sanyal, A novel approach for panic-face extraction based on mutation, in 2012 IEEE International Conference on Advanced Communication Control and Computing Technologies (ICACCCT), (2012), 473–477. https://doi.org/10.1109/ICACCCT.2012.6320825 |
[35] |
J. Lee, S. Kim, S. Kim, K. Sohn, Multi-Modal recurrent attention networks for facial expression recognition, IEEE Trans. Image Process., 29 (2020), 6977–6991. https://doi.org/10.1109/TIP.2020.2996086 doi: 10.1109/TIP.2020.2996086
![]() |
[36] |
Z. Kang, S. J. Landry, An eye movement analysis algorithm for a multielement target tracking task: Maximum transition-based agglomerative hierarchical clustering, IEEE Trans. Hmman Mach. Syst., 45 (2015), 13–24. https://doi.org/10.1109/THMS.2014.2363121 doi: 10.1109/THMS.2014.2363121
![]() |
[37] | M. A. Hossain, D. Samanta, G. Sanyal, Statistical approach for extraction of panic expression, in 2012 Fourth International Conference on Computational Intelligence and Communication Networks, (2012), 420–424. https://doi.org/10.1109/CICN.2012.189 |
[38] |
R. Janarthanan, E. A. Refaee, K. Selvakumar, M. A. Hossain, S. Rajkumar, K. Marimuthu, Biomedical image retrieval using adaptive neuro-fuzzy optimized classifier system, Math. Biosci. Eng., 19 (2022), 8132–8151. https://doi.org/10.3934/mbe.2022380 doi: 10.3934/mbe.2022380
![]() |
[39] |
F. Bu, T. Pu, W. Huang, L. Zhu, Performance and evaluation of five-phase dual random SVPWM strategy with optimized probability density function, IEEE Trans. Ind. Electron., 66 (2019), 3323–3332. https://doi.org/10.1109/TIE.2018.2854570 doi: 10.1109/TIE.2018.2854570
![]() |
[40] |
B. Manda, P. Bhaskare, R. Muthuganapathy, A convolutional neural network approach to the classification of engineering models, IEEE Access, 9 (2021), 22711–22723. https://doi.org/10.1109/ACCESS.2021.3055826 doi: 10.1109/ACCESS.2021.3055826
![]() |
[41] |
M. A. Hossain, G. Sanyal, A stochastic statistical approach for tracking human activity, IJITMC, 1 (2013), 33–42. https://doi.org/10.5121/ijitmc.2013.1304 doi: 10.5121/ijitmc.2013.1304
![]() |
[42] |
A. J. A. AlBdairi, Z. Xiao, M. Alghaili, Identifying ethnics of people through face recognition: A deep CNN approach, Sci. Prog., 2020 (2020), 6385281. https://doi.org/10.1155/2020/6385281 doi: 10.1155/2020/6385281
![]() |
[43] |
N. Alay, H. H. Al-Baity, Deep learning approach for multimodal biometric recognition system based on fusion of iris, face, and finger vein traits, Sensor, 20 (2020), 5523–5539. https://doi.org/10.3390/s20195523 doi: 10.3390/s20195523
![]() |
1. | Saleh Khoshnood, Mehdi Tavalla, Mohammad Ali Mohaghegh, Faham Khamesipour, Seyed Hossein Hejazi, Genetic diversity of Leishmania major strains isolated from Phlebotomus papatasi and Phlebotomus alexandri in western Iran based on minicircle kDNA, 2021, 0971-7196, 10.1007/s12639-021-01369-y | |
2. | Alfredo Juarez-Saldivar, Rogelio Gómez-Escobedo, Gerardo Corral-Ruiz, Karla Fabiola Chacón-Vargas, Vanessa Horta-Montaño, Luvia Sanchez-Torres, Lenci k. Vazquez-Jimenez, Benjamín Nogueda-Torres, Gildardo Rivera, Repositioning FDA-Approved Drug Against Chagas Disease and Cutaneous Leishmaniosis by Structure-Based Virtual Screening, 2024, 55, 01884409, 102958, 10.1016/j.arcmed.2024.102958 | |
3. | Andrea Lafleur, Stephane Daffis, Charles Mowbray, Byron Arana, Immunotherapeutic Strategies as Potential Treatment Options for Cutaneous Leishmaniasis, 2024, 12, 2076-393X, 1179, 10.3390/vaccines12101179 | |
4. | Anindita Bhattacharjee, Arka Bagchi, Solanki Sarkar, Sriparna Bawali, Arijit Bhattacharya, Arunima Biswas, Repurposing approved protein kinase inhibitors as potent anti-leishmanials targeting Leishmania MAP kinases, 2024, 351, 00243205, 122844, 10.1016/j.lfs.2024.122844 |
miRNA | Biofluid type | Sample size | Sample type | Method | Reference | |
↓ miR-29a | Peripheral blood White blood cells Serum | 15 PD vs. 8 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | qRT-PCR (SYBR green) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Margis et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-29c | Peripheral blood mononuclear cell White blood cell Serum | 19 PD vs.13 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-30c | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Vallelunga et al., 2014 | |
↓ miR-19b | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Botta-Orfila et al., 2014 | |
↓ miR-1, miR-22* ↑ miR-16-2*, miR-26a2*, miR-30a | Peripheral blood | 15 PD vs. 8 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD levodopa/carbidopa treated vs. CTR | qRT-PCR (SYBR green) | Margis et al., 2011 | |
↓ miR-335, miR-374a, miR-199a-3p/miR-199b-3p, miR-126*, miR-151-3p, miR-199a-5p, miR-151-5p, miR-126, miR-29b, miR-147, miR-28-5p, miR-30b, miR-374b, miR-301a, miR-26a | Peripheral blood mononuclear cell | 19 PD vs.13 CTR | PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) | Martins et al., 2011 | |
↓ miR-320a/b/c, miR-769, miR-92b, miR-16 ↑ miR-199b, miR-1274b, miR-21, miR-150, miR-671, miR-1249, miR-20a, miR-18b*, miR-378c, miR-4293 ↓ miR-4293, miR-378c, miR-18b*, miR-424*, miR-20a, miR-210, miR-1249, miR-93 ↑ miR-4317, miR-143, miR-424 ↓ miR-652, miR-15a*, miR-376c, miR-143 ↑ miR-423, miR-365, miR-486, miR-1260, miR-218 | White blood cell | 7 PD vs. 6 CTR | PD (pre-DBS) vs. CTR PD (pre-DBS) vs. PD (post-DBS) PD (DBS-off) vs PD (DBS-on) | Small RNA-sequencing (ABI SOLiD) | Soreq et al., 2013 | |
A panel of four miRNA biomarkers ↑ miR-1826, ↓ miR-450b-3p ↑ miR-626, miR-505 | Plasma | Discovery set: 32 PD vs.32 CTR Replication set: 42 PD vs.30 CTR Validation set: 30 PD, 4 MSA, 5 PSP vs. 8 CTR | PD (treated) vs. CTR | Agilent miRNA microarray v.3, validation by qRT-PCR (TaqMan) | Khoo et al., 2012 | |
↑ miR-331-5p | Plasma | 31 PD vs. 25 CTR | PD (non-treated) vs. CTR | TaqMan Low Density miRNA array (3 pools of PD and CTR), then qRT-PCR (TaqMan) | Cardo et al., 2013 | |
↓ miR-339-5p ↑ miR-223*, miR-324-3p, miR-24 ↓ miR-148b ↑ miR-223*, miR-324-3p, miR-24 | Serum | Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD + MSA vs. CTR PD vs. CTR | TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Vallelunga et al., 2014 | |
↓ miR-19a | Serum | Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR | TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Botta-Orfila et al., 2014 | |
↓ miR-16-2-3p, miR-1294 ↑ miR-338-3p, miR-30e-3p, miR-30a-3p ↓ miR-132-5p, miR-485-5p, miR-127-3p, miR-128, miR-409-3p, miR-433, miR-370, miR-431-3p, miR-873-3p, miR-136-3p, miR-212-3p, miR-10a-5p, miR-1224-5p, miR-4448 ↑ miR-19a-3p, miR-19b-3p, let-7g-3p | Serum (post mortem) CSF (post mortem) | 50 PD vs. 62 CTR 57 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Small RNA Sequencing (Illumina HiSeq2000) | Burgos et al., 2014 | |
PD: Parkinson’s disease; CTR: healthy control; PSP: progressive supranuclear palsy; MSA: multiple system atrophy; LRRK2 PD: G2019S mutation in PD; DBS: Deep brain stimulation; ↓: down-regulated miRNA expression; ↑: up-regulated miRNA expression. |
miRNA | Biofluid type | Sample size | Sample type | Method | Reference | |
↓ miR-29a | Peripheral blood White blood cells Serum | 15 PD vs. 8 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | qRT-PCR (SYBR green) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Margis et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-29c | Peripheral blood mononuclear cell White blood cell Serum | 19 PD vs.13 CTR 7 PD vs. 6 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD (DBS-off) vs. PD (DBS-on) PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) Small RNA-sequencing (ABI SOLiD) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Soreq et al., 2013 Botta-Orfila et al., 2014 | |
↓ miR-30c | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Vallelunga et al., 2014 | |
↓ miR-19b | Peripheral blood mononuclear cell Serum | 19 PD vs.13 CTR Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Martins et al., 2011 Botta-Orfila et al., 2014 | |
↓ miR-1, miR-22* ↑ miR-16-2*, miR-26a2*, miR-30a | Peripheral blood | 15 PD vs. 8 CTR | de novo PD + levodopa/carbidopa treated vs. CTR + early onset PD levodopa/carbidopa treated vs. CTR | qRT-PCR (SYBR green) | Margis et al., 2011 | |
↓ miR-335, miR-374a, miR-199a-3p/miR-199b-3p, miR-126*, miR-151-3p, miR-199a-5p, miR-151-5p, miR-126, miR-29b, miR-147, miR-28-5p, miR-30b, miR-374b, miR-301a, miR-26a | Peripheral blood mononuclear cell | 19 PD vs.13 CTR | PD vs. CTR | Exiqon miRCURY LNA array v.10.0, validation by qRT-PCR (TaqMan) | Martins et al., 2011 | |
↓ miR-320a/b/c, miR-769, miR-92b, miR-16 ↑ miR-199b, miR-1274b, miR-21, miR-150, miR-671, miR-1249, miR-20a, miR-18b*, miR-378c, miR-4293 ↓ miR-4293, miR-378c, miR-18b*, miR-424*, miR-20a, miR-210, miR-1249, miR-93 ↑ miR-4317, miR-143, miR-424 ↓ miR-652, miR-15a*, miR-376c, miR-143 ↑ miR-423, miR-365, miR-486, miR-1260, miR-218 | White blood cell | 7 PD vs. 6 CTR | PD (pre-DBS) vs. CTR PD (pre-DBS) vs. PD (post-DBS) PD (DBS-off) vs PD (DBS-on) | Small RNA-sequencing (ABI SOLiD) | Soreq et al., 2013 | |
A panel of four miRNA biomarkers ↑ miR-1826, ↓ miR-450b-3p ↑ miR-626, miR-505 | Plasma | Discovery set: 32 PD vs.32 CTR Replication set: 42 PD vs.30 CTR Validation set: 30 PD, 4 MSA, 5 PSP vs. 8 CTR | PD (treated) vs. CTR | Agilent miRNA microarray v.3, validation by qRT-PCR (TaqMan) | Khoo et al., 2012 | |
↑ miR-331-5p | Plasma | 31 PD vs. 25 CTR | PD (non-treated) vs. CTR | TaqMan Low Density miRNA array (3 pools of PD and CTR), then qRT-PCR (TaqMan) | Cardo et al., 2013 | |
↓ miR-339-5p ↑ miR-223*, miR-324-3p, miR-24 ↓ miR-148b ↑ miR-223*, miR-324-3p, miR-24 | Serum | Discovery set: 6 PD, 6 MSA-P, 3 MSA-C vs. 5 CTR Validation set: 25 PD, 25 MSA, vs. 25 CTR | PD + MSA vs. CTR PD vs. CTR | TaqMan Low Density miRNA array v.3.0 A and B, validation by qRT-PCR (TaqMan) | Vallelunga et al., 2014 | |
↓ miR-19a | Serum | Discovery set: 10 PD, 10 LRRK2 PD vs. 10 CTR Validation set 1: 20 PD, 20 LRRK2 PD vs. 20 CTR Validation set 2: 65 PD vs. 65 CTR | PD vs. CTR | TaqMan Low Density miRNA array v.2.0 A, validation by qRT-PCR (TaqMan) | Botta-Orfila et al., 2014 | |
↓ miR-16-2-3p, miR-1294 ↑ miR-338-3p, miR-30e-3p, miR-30a-3p ↓ miR-132-5p, miR-485-5p, miR-127-3p, miR-128, miR-409-3p, miR-433, miR-370, miR-431-3p, miR-873-3p, miR-136-3p, miR-212-3p, miR-10a-5p, miR-1224-5p, miR-4448 ↑ miR-19a-3p, miR-19b-3p, let-7g-3p | Serum (post mortem) CSF (post mortem) | 50 PD vs. 62 CTR 57 PD vs. 65 CTR | PD vs. CTR PD vs. CTR | Small RNA Sequencing (Illumina HiSeq2000) | Burgos et al., 2014 | |
PD: Parkinson’s disease; CTR: healthy control; PSP: progressive supranuclear palsy; MSA: multiple system atrophy; LRRK2 PD: G2019S mutation in PD; DBS: Deep brain stimulation; ↓: down-regulated miRNA expression; ↑: up-regulated miRNA expression. |