Namariq Dhahir
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Namariq Dhahir has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Amer A. AbuGhazaleh
School of Agricultural Sciences, Southern Illinois University, Carbondale, IL 62901, USA
3 shared publications - Anas Alsobeh
Department of Information Systems and Technology, Utah Valley University, Orem, UT 84058, USA
2 shared publications - Katherine A. Thompson-Witrick
Department of Food Science and Human Nutrition, University of Florida, Gainesville, FL 32611, USA
2 shared publications - Malek Rababa
School of Computing, Southern Illinois University, Carbondale, IL 62901, USA
2 shared publications Amer AbuGhazaleh
Department of Animal Science, Food and Nutrition, Southern Illinois University, Carbondale, USA
1 shared publicationHasan Shamimul
1 shared publication
Latest Publications
- Open Access
- 23 October 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (11) , 525
BactSegNet: A Growth-Stage-Aware, Searched Instance-Segmentation Model for Salmonella and Escherichia coli Microcolonies in Pure-Culture Bright-Field Microscopy with a Food-Matrix Surrogate
Instance segmentation of small, dense, low-contrast microcolonies against confounding food matrices limits early microscopy-based screening for Salmonella and E. coli. We present BactSegNet, combining EfficientNetV2, BiFPN, selective-kernel attention, and growth-stage FiLM conditioning based on a learnable embedding of recorded acquisition stage. BE-NAS explored a 2304-configuration search space and evaluated 71 unique architectures. Across twelve retrained candidates, proxy and full-training rankings agreed (Spearman for mAP@50 and for mAP@50–95). Across the 71 architectures evaluated in the BE-NAS proxy search, three-seed proxy fitness showed high reliability (; 213 seed-level evaluations), while phenotype caching reduced 180 scheduled evaluations to 71 unique trainings (213 rather than 540 seed-level proxy runs). On expert-verified polygon instances spanning six stages, two species, and onion-present/absent conditions, BactSegNet-L achieved mask mAP@50 = and mAP@50–95 = , percentage points above a multi-seed YOLOv8 baseline and higher mAP@50 than sixteen class-aware single-run references across three paradigms. A capacity-preserving stage-perturbation control reduced mAP@50 from with correct stage assignments to after shuffling image–stage correspondence across three seeds. Gains reached pp for Salmonella at h and pp on onion. Across 2756 matched instances, area MAPE was with , dominated by the largest instances. BactSegNet-Efficient retained mAP@50 with fewer FLOPs at fps. The main dataset comprises pure-culture frames; an independent, expert-annotated mixed-culture evaluation (400 images, 3.5 and 4.0 h) yielded mask mAP@50= across three seeds without retraining. Recorded growth-stage metadata are required at inference, unlike the image-only comparison models. These results support early, physically calibrated microcolony segmentation while distinguishing the original pure-culture evaluation from the additional mixed-culture test.- 23 October 2026
- Published by MDPI AG in Journal of Imaging
- Vol. 12 (11) , 525
Open AccessBactSegNet: A Growth-Stage-Aware, Searched Instance-Segmentation Model for Salmonella and Escherichia coli Microcolonies in Pure-Culture Bright-Field Microscopy with a Food-Matrix Surrogate
Instance segmentation of small, dense, low-contrast microcolonies against confounding food matrices limits early microscopy-based screening for Salmonella and E. coli. We present BactSegNet, combining EfficientNetV2, BiFPN, selective-kernel attention, and growth-stage FiLM conditioning based on a learnable embedding of recorded acquisition stage. BE-NAS explored a 2304-configuration search space and evaluated 71 unique architectures. Across twelve retrained candidates, proxy and full-training rankings agreed (Spearman for mAP@50 and for mAP@50–95). Across the 71 architectures evaluated in the BE-NAS proxy search, three-seed proxy fitness showed high reliability (; 213 seed-level evaluations), while phenotype caching reduced 180 scheduled evaluations to 71 unique trainings (213 rather than 540 seed-level proxy runs). On expert-verified polygon instances spanning six stages, two species, and onion-present/absent conditions, BactSegNet-L achieved mask mAP@50 = and mAP@50–95 = , percentage points above a multi-seed YOLOv8 baseline and higher mAP@50 than sixteen class-aware single-run references across three paradigms. A capacity-preserving stage-perturbation control reduced mAP@50 from with correct stage assignments to after shuffling image–stage correspondence across three seeds. Gains reached pp for Salmonella at h and pp on onion. Across 2756 matched instances, area MAPE was with , dominated by the largest instances. BactSegNet-Efficient retained mAP@50 with fewer FLOPs at fps. The main dataset comprises pure-culture frames; an independent, expert-annotated mixed-culture evaluation (400 images, 3.5 and 4.0 h) yielded mask mAP@50= across three seeds without retraining. Recorded growth-stage metadata are required at inference, unlike the image-only comparison models. These results support early, physically calibrated microcolony segmentation while distinguishing the original pure-culture evaluation from the additional mixed-culture test. - Open Access
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 112-119
XAIPath: Temporal-Environmental Explainable AI Framework for Co-Contaminated Food Pathogen Detection in Microscopic Imaging
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 112-119
Open AccessXAIPath: Temporal-Environmental Explainable AI Framework for Co-Contaminated Food Pathogen Detection in Microscopic Imaging
- Open Access
- 22 January 2021
- Published by Science and Education Publishing Co., Ltd. in American Journal of Food Science and Technology
- Vol. 9 (4) , 134-141
Camel Milk Composition and Microbial Reduction with Different Pasteurization Methods
The objective of this study was to investigate the efficacy of different thermal pasteurization methods on (1) the survival of the total aerobic bacteria, E. coli O157: H7, in camel milk, and (2) the camel milk components such as the fatty acid profile, lipid peroxidation, protein fractions, and the composition of volatile compounds. Samples of camel milk (N=9) were pasteurized at 65°C/30 min (PAST-1), 72°C/5 min (PAST-2), and 80°C/15 min (PAST-3). The survival of E. coli O157: H7 was evaluated using the traditional plate count agar (PCA) method while the total aerobic bacteria were enumerated using the petrifilm aerobic count plates (ACP). Complete elimination (P10 CFU/ml reduction). All pasteurization methods had a significant (PE. coli O157: H7 resulting in a 6 log10 CFU/ml reduction. There were no significant (P>0.05) differences in the fatty acid profile including the cis-9, trans-11 conjugated linoleic acid (CLA), and trans-10, cis-12 CLA, and the lipid peroxidation products between raw and pasteurized milk samples. The milk protein profile was marginally altered by PAST-2 and PAST-3 treatments but not PAST-1. Thirty-four volatile compounds (VCs) were detected in the raw milk samples compared to 29 VCs in the pasteurized milk samples. Pasteurization treatments altered the concentrations of some milk VCs, increasing the Heptanal, Tridecanal, and Undecanal while decreasing the 2-Decanal and 2-Undecanal. This study shows that PAST-1 and PAST-3 treatments are more effective than PAST-2 at inactivating total aerobic bacteria. Additionally, the absence of significant changes in milk compositions indicates that PAST-1 and PAST-3 could be applied without affecting the nutritional value of camel milk.
- 22 January 2021
- Published by Science and Education Publishing Co., Ltd. in American Journal of Food Science and Technology
- Vol. 9 (4) , 134-141
Open AccessCamel Milk Composition and Microbial Reduction with Different Pasteurization Methods
The objective of this study was to investigate the efficacy of different thermal pasteurization methods on (1) the survival of the total aerobic bacteria, E. coli O157: H7, in camel milk, and (2) the camel milk components such as the fatty acid profile, lipid peroxidation, protein fractions, and the composition of volatile compounds. Samples of camel milk (N=9) were pasteurized at 65°C/30 min (PAST-1), 72°C/5 min (PAST-2), and 80°C/15 min (PAST-3). The survival of E. coli O157: H7 was evaluated using the traditional plate count agar (PCA) method while the total aerobic bacteria were enumerated using the petrifilm aerobic count plates (ACP). Complete elimination (P10 CFU/ml reduction). All pasteurization methods had a significant (PE. coli O157: H7 resulting in a 6 log10 CFU/ml reduction. There were no significant (P>0.05) differences in the fatty acid profile including the cis-9, trans-11 conjugated linoleic acid (CLA), and trans-10, cis-12 CLA, and the lipid peroxidation products between raw and pasteurized milk samples. The milk protein profile was marginally altered by PAST-2 and PAST-3 treatments but not PAST-1. Thirty-four volatile compounds (VCs) were detected in the raw milk samples compared to 29 VCs in the pasteurized milk samples. Pasteurization treatments altered the concentrations of some milk VCs, increasing the Heptanal, Tridecanal, and Undecanal while decreasing the 2-Decanal and 2-Undecanal. This study shows that PAST-1 and PAST-3 treatments are more effective than PAST-2 at inactivating total aerobic bacteria. Additionally, the absence of significant changes in milk compositions indicates that PAST-1 and PAST-3 could be applied without affecting the nutritional value of camel milk.
- 24 June 2020
- Published by SAGE Publications in Food Science and Technology International
- Vol. 27 (2) , 99-111
The effect of different ultraviolet-C light doses on microbial reduction and the components of camel milk
As a result of increasing interest in non-thermal technologies as a possible alternative or complementary to milk pasteurization processing, the objectives of this study were to determine the effects of different ultraviolet-C light doses on the viability of Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium and chemical changes to camel milk components. Pasteurized and inoculated camel milk samples were ultraviolet-C treated in a continuous flow system. The viability of E. coli O157:H7 and S. Typhimurium was evaluated with both in vivo imaging system and traditional plate count agar method. Samples subjected to the 4.15, 8.30, and 12.45 mJ/cm2 of ultraviolet-C treatment resulted in 1.9, 3.3, and 3.9-log reductions in E. coli O157:H7 and 0.9, 3, and 3.9-log reductions in S. Typhimurium, respectively. The measurement of secondary lipid peroxidation products (or ThioBarbituric Acid Reactive Substance values) showed no significant (P > 0.05) differences between the raw and ultraviolet-C treated milk samples. Additionally, no changes (P > 0.05) in the protein profiles of αs1-casein, α-lactalbumin, and lactoferrin were observed between both samples. Compared to the untreated raw milk, c9 t11 conjugated linoleic acid decreased (P < 0.01) while t10 c12 conjugated linoleic acid increased (P < 0.01) in the ultraviolet-C treated milk. Furthermore, three new volatile compounds were identified in the ultraviolet-C treated milk compared to the control. In conclusion, milk treated with the ultraviolet-C light at a dose of 12.45 mJ/cm2 did not meet the Food and Drug Administration (FDA) requirements for the 5-log pathogen reduction. The ultraviolet-C treatment, on the other hand, had minimal effects on camel milk components.
- 24 June 2020
- Published by SAGE Publications in Food Science and Technology International
- Vol. 27 (2) , 99-111
The effect of different ultraviolet-C light doses on microbial reduction and the components of camel milk
As a result of increasing interest in non-thermal technologies as a possible alternative or complementary to milk pasteurization processing, the objectives of this study were to determine the effects of different ultraviolet-C light doses on the viability of Escherichia coli O157:H7 and Salmonella enterica serovar Typhimurium and chemical changes to camel milk components. Pasteurized and inoculated camel milk samples were ultraviolet-C treated in a continuous flow system. The viability of E. coli O157:H7 and S. Typhimurium was evaluated with both in vivo imaging system and traditional plate count agar method. Samples subjected to the 4.15, 8.30, and 12.45 mJ/cm2 of ultraviolet-C treatment resulted in 1.9, 3.3, and 3.9-log reductions in E. coli O157:H7 and 0.9, 3, and 3.9-log reductions in S. Typhimurium, respectively. The measurement of secondary lipid peroxidation products (or ThioBarbituric Acid Reactive Substance values) showed no significant (P > 0.05) differences between the raw and ultraviolet-C treated milk samples. Additionally, no changes (P > 0.05) in the protein profiles of αs1-casein, α-lactalbumin, and lactoferrin were observed between both samples. Compared to the untreated raw milk, c9 t11 conjugated linoleic acid decreased (P < 0.01) while t10 c12 conjugated linoleic acid increased (P < 0.01) in the ultraviolet-C treated milk. Furthermore, three new volatile compounds were identified in the ultraviolet-C treated milk compared to the control. In conclusion, milk treated with the ultraviolet-C light at a dose of 12.45 mJ/cm2 did not meet the Food and Drug Administration (FDA) requirements for the 5-log pathogen reduction. The ultraviolet-C treatment, on the other hand, had minimal effects on camel milk components.