Accuracy of angioboost software versus routine angiography for assessing the degree of coronary artery lesion occlusion

Abstract

Background:

Accurate image extraction is essential for simplifying the diagnosis and treatment of coronary artery disease (CAD) and reducing the need for multiple imaging procedures. This study compared the diagnostic accuracy of AngioBoost knowledge-based software with routine angiography in assessing coronary artery lesions.

Methods:

This cross-sectional study included 174 patients undergoing coronary angiography at Shahid Rajaei Heart Hospital, Tehran. Patients with signs and symptoms of ischemic heart disease and an indication for coronary angiography were enrolled. All underwent angiography using the Seldinger method. Angiographic films were reviewed by two independent cardiologists. AngioBoost was then applied to the same images, and its findings were interpreted by two additional cardiologists.

Results:

Comparison of routine angiography and AngioBoost showed contingency coefficients of 74.7% (P = 0.01) and 90.1% (P = 0.01) for the first and second cardiologists, respectively. AngioBoost correctly identified 96.3% of patent vessels, 79.9% of mild lesions, 58.0% of moderate lesions, 91.2% of significant lesions, and 96.8% of occluded vessels. Overall diagnostic agreement was 88.2% (P = 0.01). Kappa coefficients for vascular occlusion were 46.8% for Left Main, 71.7% for LAD, 70.1% for LCX, and 77.8% for RCA (all P = 0.01).

Conclusion:

AngioBoost demonstrated acceptable diagnostic accuracy in assessing coronary artery occlusion, showing strong agreement (88.2%) with routine angiography. These findings suggest that AngioBoost may serve as a useful complementary tool for CAD diagnosis, potentially reducing physician workload and improving diagnostic efficiency.

Cardiovascular disease (CVD) remains the leading cause of disability and mortality worldwide, including in Iran (1). It significantly impacts an individual's ability to maintain normal function and imposes a major burden on healthcare systems. In Iran, the prevalence of CVD is estimated at 3,500 cases per 100,000 people(2), with a worrying increase from 5.7% to 17.8% in recent years. About 50% of death in Iran is caused by coronary artery disease, which has the highest death rate among other diseases (3). CAD occurs primarily due to the obstruction of myocardial blood flow, leading to an inadequate oxygen and nutrient supply to the heart tissue. This obstruction results from excessive accumulation of atherosclerotic plaques and fatty deposits within the coronary arteries. The rupture of these plaques can trigger clot formation, further narrowing or completely blocking the arteries, leading to acute cardiovascular events such as myocardial infarction (4).Early detection of coronary artery lesions is crucial, as it provides valuable prognostic information and allows for timely intervention, potentially preventing heart failure progression (5).

Several diagnostic imaging modalities are available for assessing coronary artery involvement, including computed tomographic angiography (CTA), magnetic resonance angiography (MRA), and x-ray coronary angiography (XCA). More advanced techniques, such as intravascular ultrasound (IVUS) and optical coherence tomography (OCT), provide detailed insights into vessel morphology and plaque composition (6, 7). Despite these advancements, x-ray coronary angiography remains the gold standard for diagnosing CAD, given its cost-effectiveness, accessibility, and superior temporal and spatial resolution compared to other imaging methods (8). However, its accuracy can be affected by image noise, motion artifacts, and background complexity (9).

To improve the precision of coronary artery lesion assessment, vessel extraction techniques are often employed to enhance image quality by reducing background noise and isolating vessel structures. Enhancing angiographic images not only aids in better visualization of lesions but also minimizes the number of images required for diagnosis, thereby optimizing the diagnosis (10). Recent advancements in artificial intelligence and image-processing algorithms have led to improved diagnostic tools, such as Angioboost. This knowledge-based software is specifically designed to enhance coronary angiographic images. By utilizing advanced image enhancement techniques, Angioboost aimed to improve the detection and classification of lesions, potentially offering greater accuracy than conventional angiography.

AngioBoost is a knowledge-based image enhancement software developed by the Biomedical Engineering Department of Amirkabir University of Technology (Tehran Polytechnic), Iran. The software aimed to improve the visualization of coronary arteries in routine angiography by enhancing vessel edges and reducing background noise. It is currently available for clinical research use in selected imaging and cardiology centers in Iran, under institutional licensing, and is not yet distributed as a public or commercial software. Its main competitive advantages include compatibility with standard angiography equipment, independence from proprietary systems, and cost-effectiveness for use in both clinical and academic centers. Additionally, it enables more precise vessel visualization without increasing imaging time or radiation exposure, making it a practical diagnostic aid in routine cardiology settings.

Therefore, this study aimed to compare the accuracy of Angioboost software with routine x-ray angiography in assessing the degree of coronary artery lesion occlusion. The findings of this study could provide valuable insights into whether advanced software-based analysis can complement or even enhance conventional angiographic assessments, ultimately contributing to more accurate and efficient CAD diagnosis and treatment.

Methods

Study population:

This cross-sectional study included 174 patients who were candidates for diagnostic coronary angiography at Shahid Rajaei Heart Hospital, Tehran, in 2022. The inclusion criteria were based on the clinical guidelines for the evaluation of ischemic heart disease and the indications for coronary angiography established by the American College of Cardiology (ACC) and the American Heart Association (AHA) (11-13).

These indications included the presence of angina pectoris or equivalent symptoms, positive non-invasive ischemia tests, abnormal ECG findings, and high clinical suspicion of coronary artery disease unconfirmed by other imaging modalities. Patients with contraindications to angiography or incomplete medical records were excluded. The diagnostic criteria for classifying the degree of coronary artery occlusion (patent, mild, moderate, significant, and cut) were also defined according to standard quantitative coronary angiography (QCA) criteria, consistent with previous studies.

Angiography protocol:

Following the standard angiographic procedure, the AngioBoost software was employed to analyze the patients' angiographic films. The software-generated results were independently evaluated by two additional cardiologists. To ensure objectivity, all cardiologists were blinded to both the expert interpretations and the software-generated findings. The primary variable assessed in this study was the degree of coronary artery occlusion, categorized as follows:

- Patent

- Mild occlusion

- Moderate occlusion

- Significant occlusion

- Complete occlusion (cut)

Coronary lesions were evaluated across multiple arterial segments and branches, including:

Left Coronary Artery (LCA):

- Left Main Ostial part (LMO)

- Left Main Distal part (LMD)

- Left Anterior Descending Ostial part (LADO)

- Left Anterior Descending Proximal part (LADP)

- Left Anterior Descending Mid part (LADM)

- Left Anterior Descending Distal part (LADD)

- Left Circumflex Ostial part (LCXO)

- Left Circumflex Proximal part (LCXP)

- Left Circumflex Mid part (LCXM)

- Left Circumflex Distal part (LCXD)

Right Coronary Artery (RCA):

- Right Coronary Artery Ostial part (RCAO)

- Right Coronary Artery Proximal part (RCAP)

- Right Coronary Artery Mid part (RCAM)

- Right Coronary Artery Distal part (RCAD)

The AngioBoost software (Version 1.2, Amirkabir University of Technology, Tehran, Iran) was installed and operated in the imaging unit of Shahid Rajaei Heart Hospital. The software runs on standard Windows-based workstations and is specifically designed for post-processing of routine X-ray angiography images. Access is limited to authorized medical and research centers through institutional collaboration agreements.

Statistical analysis:

All data were analyzed using SPSS version 26. The Chi-square test was used to compare the frequency distribution of occlusion categories between the AngioBoost software and routine angiography interpretations. Additionally, the Kappa agreement coefficient was calculated to assess the inter-rater agreement between the two methods. A p-value of <0.05 was considered statistically significant.

Results

Table 1 presents the findings of the first physician's observations using two diagnostic methods: routine angiography and Angioboost software. The results indicate a high level of agreement between the two methods in diagnosing vascular occlusion, with agreement rates ranging from 82.4% in LCXM to 99.4% in LADO, demonstrating a strong correlation between the diagnostic approaches.

Table 1.

Frequency distribution and agreement of the first physician's observations using routine angiography and Angioboost

Kappa	P-value	Angioboost					Standard angiography
		Cut	Significant	Moderate	Mild	Patent
62.6	0.001>	0	0	0	1 (0.6)	167 (99.4)	Patent
		0	1 (16.7)	0	3 (50.0)	2 (33.3)	Mild
63.7	0.001>	0	2 (3.1)	0	4 (2.5)	151 (96.2)	Patent
		0	1 (3.8)	0	8 (66.7)	3 (25.0)	Mild
		0	3 (60.0)	0	1 (20.0)	1 (20.0)	Significant
85.8	0.001>	0	0	0	1 (0.6)	158 (99.4)	Patent
		0	0	0	1 (50.0)	1 (50.0)	Mild
		0	0	0	2 (100)	0	Moderate
		0	9 (100)	0	0	0	Significant
		2 (100)	0	0	0	0	Cut
80.0	0.001>	0	0	1 (0.8)	2 (1.7)	115 (97.5)	Patent
		0	0	1 (4.3)	22 (75.9)	6 (20.7)	Mild
		0	0	3 (50.0)	2 (33.3)	1 (16.7)	Moderate
		0	15 (78.9)	3 (15.8)	1 (5.3)	0	Significant
		2 (100)	0	0	0	0	Cut
75.3	0.001>	0	0	0	8 (5.5)	137 (94.5)	Patent
		0	0	0	19 (79.2)	4 (16.7)	Mild
		0	0	0	1 (100)	0	Moderate
		0	4 (100)	0	0	0	Significant
		1 (100)	0	0	0	0	Cut
72.1	0.001>	0	0	0	10 (11.1)	80 (87.9)	Patent
		0	0	0	42 (73.7)	11 (19.3)	Mild
		0	0	2 (40.0)	1 (20.0)	1 (20.0)	Moderate
		0	28 (82.4)	2 (5.9)	2 (5.9)	0	Significant
		9 (90.0)	1 (10.0)	0	0	0	Cut
68.1	0.001>	0	0	0	13 (6.2)	61 (82.4)	Patent
		0	0	0	44 (71.0)	16 (25.8)	Mild
		0	0	0	2 (25.0)	3 (37.5)	Moderate
		0	23 (95.8)	1 (4.2)	0	0	Significant
		6 (100)	0	0	0	0	Cut
67.5	0.001>	0	0	0	2 (2.7)	8 (90.4)	Patent
		0	0	0	39 (75.0)	12 (23.1)	Mild
		0	0	1 (100)	0	0	Moderate
		0	5 (71.4)	1 (14.3)	1 (14.3)	0	Significant
		1 (100)	0	0	0	0	Cut
88.7	0.001>	0	0	0	1 (0.6)	166 (99.4)	Patent
		0	0	0	2 (100)	0	Mild
		0	1 (100)	0	0	0	Significant
		1 (100)	0	0	0	0	Cut
75.7	0.001>	0	0	0	7 (2.7)	90 (92.8)	Patent
		0	0	0	39 (70.9)	14 (25.5)	Mild
		0	1 (25.0)	3 (75.0)	0	0	Moderate
		0	14 (100)	0	0	0	Significant
		3 (100)	0	0	0	0	Cut
69.6	0.001>	0	0	0	2 (2.2)	81 (89.0)	Patent
		0	0	0	42 (73.7)	11 (19.3)	Mild
		0	0	2 (50.0)	1 (25.0)	1 (25.0)	Moderate
		0	12 (85.7)	2 (14.3)	0	0	Significant
		5 (83.3)	1 (16.7)	0	0	0	Cut
67.8	0.001>	0	0	0	2 (2.8)	11 (97.2)	Patent
		0	0	0	41 (77.4)	10 (18.9)	Mild
		0	0	1 (100)	0	0	Moderate
		0	4 (80.0)	1 (20.0)	0	0	Significant

LMO -- Left Main Ostial; LMD -- Left Main Distal; LADO -- Left Anterior Descending Ostial; LADP -- Left Anterior Descending Proximal; LADM -- Left Anterior Descending Mid; LADD -- Left Anterior Descending Distal; LCXO -- Left Circumflex Ostial; LCXP -- Left Circumflex Proximal; LCXM -- Left Circumflex Mid; LCXD-- Left Circumflex Distal; RCAO -- Right Coronary Artery Ostial; RCAP -- Right Coronary Artery Proximal; RCAM -- Right Coronary Artery Mid; RCAD -- Right Coronary Artery Distal; Kappa -- Statistical measure of inter-rater agreement; p-value -- Probability value (significance level); AngioBoost -- Knowledge-based software for coronary angiography enhancement.

Notably, the agreement percentage was higher in cases of moderate, significant, and cut occlusion, whereas in cases of mild occlusion, the level of agreement was comparatively lower. This suggests that the differentiation of less severe occlusions may be more challenging, potentially due to subtle variations in imaging interpretation. Furthermore, the Kappa coefficient analysis revealed the lowest agreement level (62.6%) in diagnosing LMO using routine angiography and Angioboost, while the highest agreement (88.7%) was observed in RCAO (P < 0.001).

These findings highlight the reliability and consistency of Angioboost in most cases, particularly for more severe occlusions.

Table 2.

Frequency distribution and agreement of the second physician's observations using routine angiography and Angioboost

Kappa	P-value	Angioboost					Standard angiography
		Cut	Significant	Moderate	Mild	Patent
56.4	0.001>	0	0	0	0	169 (100)	Patent
		0	0	0	2 (40.0)	3 (60.0)	Mild
82.7	0.001>	0	0	0	2 (1.3)	156 (98.7)	Patent
		0	0	0	8 (72.7)	3 (27.3)	Mild
		0	5 (100)	0	0	0	Significant
92.4	0.001>	0	0	0	0	160 (100)	Patent
		0	0	2 (40.0)	3 (60.0)	0	Mild
		0	7 (100)	0	0	0	Moderate
		2 (100)	0	0	0	0	Significant
		0	0	0	6 (1.5)	112 (94.9)	Cut
86.6	0.001>	0	0	1 (4.3)	1 (4.3)	27 (91.3)	Patent
		0	0	5 (71.4)	2 (28.6)	0	Mild
		0	17 (89.5)	2 (10.5)	0	0	Moderate
		1 (100)	0	0	0	0	Significant
		0	0	0	0	77 (100)	Cut
86.6	0.001>	0	0	2 (7.5)	4 (11.4)	29 (82.9)	Patent
		0	2 (22.2)	7 (77.8)	0	0	Mild
		0	45 (97.8)	1 (2.2)	0	0	Moderate
		7 (100)	0	0	0	0	Significant
		0	0	0	1 (7.0)	146 (99.3)	Cut
95.8	0.001>	0	0	0	0	21 (100)	Patent
		0	0	1 (50.0)	1 (50.0)	0	Mild
		0	3 (100)	0	0	0	Significant
		1 (100)	0	0	0	0	Cut
68.4	0.001>	0	0	0	2 (1.2)	166 (98.8)	Patent
		0	0	0	0	1 (100)	Mild
		0	3 (75.0)	1 (25.0)	0	0	Significant
		1 (100)	0	0	0	0	Cut
90.7	0.001>	0	0	0	3 (2.8)	108 (97.3)	Patent
		0	0	0	49 (92.5)	4 (7.5)	Mild
		0	0	3 (100)	0	0	Moderate
		0	5 (83.3)	1 (16.7)	0	0	Significant
		1 (100)	0	0	0	0	Cut
86.6	0.001>	0	0	0	3 (4.3)	86 (95.6)	Patent
		0	0	3 (16.7)	42 (85.0)	4 (8.0)	Mild
		0	4 (38.0)	8 (56.0)	1 (7.0)	0	Moderate
		0	19 (100)	0	0	0	Significant
		4 (100)	0	0	0	0	Cut
87.3	0.001>	0	0	0	4 (1.3)	127 (98.4)	Patent
		0	0	0	35 (89.7)	4 (10.3)	Mild
		0	6 (85.7)	1 (14.3)	0	0	Moderate
		0	0	0	0	171 (100)	Significant
83.1	0.001>	0	1 (50.0)	1 (50.0)	0	0	Patent
		1 (100)	0	0	0	0	Mild
		0	0	0	5 (8.4)	100 (90.9)	Significant
		0	0	1 (2.1)	41 (87.2)	5 (10.6)	Cut
88.6	0.001>	0	0	6 (100)	0	0	Patent
		0	14 (100)	0	0	0	Mild
		3 (100)	0	0	0	0	Moderate
		0	0	0	4 (4.1)	93 (95.9)	Significant
		0	0	1 (1.2)	43 (91.5)	3 (6.4)	Cut
91.5	0.001>	0	0	6 (85.7)	1 (14.3)	0	Patent
		0	15 (100)	0	0	0	Mild
		8 (100)	0	0	0	0	Moderate
		0	0	0	2 (1.7)	118 (98.3)	Significant
		0	0	0	40 (87.0)	6 (13.0)	Cut
89.6	0.001>	0	0	3 (100)	0	0	Patent
		0	5 (100)	0	0	0	Mild
		0	1 (100)	0	0	0	Moderate
		0	4 (80.0)	1 (20.0)	0	0	Significant

LMO -- Left Main Ostial; LMD -- Left Main Distal; LADO -- Left Anterior Descending Ostial; LADP -- Left Anterior Descending Proximal; LADM -- Left Anterior Descending Mid; LADD -- Left Anterior Descending Distal; LCXO -- Left Circumflex Ostial; LCXP -- Left Circumflex Proximal; LCXM -- Left Circumflex Mid; LCXD-- Left Circumflex Distal; RCAO -- Right Coronary Artery Ostial; RCAP -- Right Coronary Artery Proximal; RCAM -- Right Coronary Artery Mid; RCAD -- Right Coronary Artery Distal; Kappa -- Statistical measure of inter-rater agreement; p-value -- Probability value (significance level); AngioBoost -- Knowledge-based software for coronary angiography enhancement.

Figure 1.

Angiography performed using the standard angiography system (upper row) was compared with the AngioBoost system (lower row), assessing the visualization of the RCA in terms of region, stent placement, restenosis, and stent positioning, with comparable results between the two methods.

Discussion

This study aimed to compare the diagnostic accuracy of AngioBoost software with conventional angiography in assessing the degree of coronary artery occlusion. A total of 174 patients underwent routine angiography, and their results were analyzed using both conventional angiography and Angioboost software by two independent cardiologists. The results demonstrated a high level of agreement between the two methods, with an overall Kappa coefficient of 88.2%, reflecting very good diagnostic consistency. These findings indicate that AngioBoost can reliably assist clinicians in the interpretation of angiographic data by enhancing vessel visibility and reducing background interference. The diagnostic performance of AngioBoost showing over 90% agreement in identifying significant and cut lesions suggests that the software can provide accuracy comparable to established quantitative coronary angiography (QCA) methods, but with lower operational complexity and faster image analysis (14, 15).

Previous studies have also explored the role of image-processing algorithms and computer-assisted systems in improving coronary lesion detection. For instance, Kiani et al. (16) and Salehi et al. (17) reported similar success in identifying coronary arteries and stenosis points using automated extraction techniques. However, unlike these methods, AngioBoost integrates both enhancement and diagnostic classification within a single framework, making it more practical for clinical use.

Compared with conventional QCA software such as CAAS and QAngio, which require manual delineation of vessel borders and operator training, AngioBoost's fully automated enhancement process reduces observer variability. The results of this study confirm that, while routine angiography remains the gold standard for coronary evaluation, post-processing with AngioBoost can significantly improve image interpretability, potentially minimizing diagnostic errors and the need for repeated imaging (18). The contingency coefficients were calculated to assess the agreement between the two diagnostic methods. For the first physician, the contingency coefficient ranged from 62.6% for LMO to 88.7% for RCAO, while for the second physician, the values ranged from 56.4% for LMO to 95.8% for LADD. When comparing the observations of both physicians, the lowest agreement (46.8%) was observed in the LMO+LMD vessels, while a higher level of agreement was seen in other vascular regions, including LADO+LADP+LADM+LADD (71.7%), LCXO+LCXP+LCXM+LCXD (71.0%), and LCAO+RCAP+RCAM+RCAD (77.8%).The overall Kappa agreement coefficient for the first and second physicians' observations using routine angiography and Angioboost was 74.7% and 90.1%, respectively. Additionally, the total diagnostic contingency across both methods was 88.2%, indicating a very high level of agreement. Specifically, agreement rates for patent vessels, mild lesions, moderate lesions, significant lesions, and completely occluded vessels were 96.3%, 79.9%, 58.0%, 91.2%, and 96.8%, respectively. Notably, the contingency coefficient for observations made using routine angiography was 80.3%, while with Angioboost, it was slightly lower at 64.3%.

Several studies have explored different diagnostic methods for assessing coronary artery occlusion. Abedi et al. reported an 84% diagnostic accuracy for myocardial perfusion scanning in detecting coronary artery disease (CAD) compared to angiography, highlighting its effectiveness as a non-invasive method (19). Similarly, Johansen et al. found an 88% accuracy rate for myocardial perfusion scans in CAD detection (20). Daghighi et al. examined the diagnostic accuracy of 64-slice computed tomography (CT) compared to invasive coronary angiography (ICA), finding correlation rates of 86% in the RCA, 77% in the LM, 74% in the LAD, and 78% in the LCA, demonstrating its high diagnostic capability. However, it remains unable to fully replace invasive angiography (21). Hosseini et al. compared multislice 64 CT angiography with conventional angiography in detecting ≥50% stenosis in coronary graft vessels, demonstrating its high accuracy in evaluating venous and arterial grafts, thereby reducing the need for conventional angiography in post-CABG patients (22).

Parizad et al. examined the impact of comparative statistical iterative reconstruction methods on CT coronary angiography image quality and radiation dose reduction, showing that the ASiR method significantly lowered radiation exposure without compromising image quality (23). Other studies have also introduced machine learning and AI-based approaches for improving diagnostic accuracy, such as Kiani et al.'s phasic average classification algorithm, which accurately identified artery width, occlusion location, intensity, and blood flow speed (16), and Salehi et al.'s tree-tracing algorithm, which demonstrated superior performance in detecting coronary artery centers in cardiac angiography images (17). Over the past three decades, significant advancements in medical imaging technology have contributed to more accurate and efficient diagnostic methods (24). However, the high cost and limited accessibility of certain advanced technologies present challenges to their widespread adoption (25). Not all modern imaging techniques offer substantial improvements over existing methods, and some lack strong evidence supporting better patient outcomes (26). Therefore, before integrating new diagnostic technologies, thorough safety and performance evaluations must be conducted to validate their clinical effectiveness.

The present study evaluated the diagnostic potential of Angioboost software, developed by researchers at Amir Kabir University, against routine angiography as the standard method. The findings demonstrated that Angioboost provides acceptable diagnostic accuracy and high contingency levels, suggesting its potential superiority over some foreign alternatives. While improvements in image quality and algorithmic performance are still needed, the software has shown promising clinical applications and could serve as a valuable tool for assisting physicians in CAD diagnosis. The AngioBoost software processes raw digital image data acquired directly from standard x-ray angiography systems (DICOM format) used during routine diagnostic procedures. No separate imaging sessions or external datasets are required. The software applies knowledge-based algorithms to improve vessel contrast, suppress background motion artifacts, and highlight lumen boundaries, allowing for clearer visualization of occlusions and stenosis.It is important to emphasize that AngioBoost does not replace routine angiography as an imaging modality; rather, it enhances the diagnostic value of existing angiograms. By improving image clarity and interpretability, it may reduce the need for additional angiographic projections, shorten analysis time, and improve inter-observer consistency. In the long term, such post-processing tools could lower procedural costs and patient exposure to contrast agents and radiation, particularly in centers lacking access to advanced quantitative coronary analysis (QCA) systems. Overall, the present results demonstrate that AngioBoost achieves diagnostic performance consistent with global benchmarks for computer-assisted angiographic interpretation, while offering greater accessibility and lower cost for regional cardiology centers.

Despite these promising results, the study has several limitations. The sample size of 174 patients, though informative, may not fully represent broader patient populations, and larger, multi-center studies are necessary for further validation. Observer variability was also evident, as the agreement levels varied between the two physicians, suggesting that experience and interpretation differences may influence diagnostic outcomes. The study also focused on diagnostic accuracy without assessing long-term patient outcomes, making it necessary to conduct follow-up studies evaluating patient prognosis and treatment efficacy based on Angioboost-assisted diagnoses. Furthermore, while this study compared routine angiography and Angioboost, it did not include comparisons with other advanced imaging modalities such as CT angiography, MRI angiography, or IVUS, which could provide a more comprehensive evaluation of Angioboost's performance.

The results suggest that Angioboost software demonstrates strong agreement with routine angiography in diagnosing coronary artery occlusion. While routine angiography remains the gold standard, Angioboost has shown significant potential as a complementary diagnostic tool. Further technological refinements, larger-scale studies, and multi-center trials are needed to optimize its performance and clinical applicability. If further developed, Angioboost could serve as a reliable and cost-effective alternative in CAD diagnosis, improving diagnostic efficiency and reducing physician workload while maintaining high accuracy in lesion detection.In summary, AngioBoost operates on the same angiographic data obtained during routine procedures, requiring no extra imaging or patient intervention. While it cannot fully substitute invasive angiography, its integration into the diagnostic workflow can significantly enhance the interpretation of angiograms by optimizing vessel visibility and diagnostic precision. Therefore, AngioBoost should be viewed as a complementary diagnostic aid that strengthens routine angiographic assessment rather than a replacement technique.