BINARY DISTRIBUTION MODEL FOR IDENTIFICATION OF PATHOLOGY IN RADIOLOGICAL IMAGES
Fuente:
Zenodo
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Recurso digital |
| Lingua: | inglese |
| Pubblicazione: |
Zenodo
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866901851011547136 |
|---|---|
| author | Umair Masood Awan |
| author_facet | Umair Masood Awan |
| contents | <p><span>Identifying pathology in radiological images involves a mathematical and computational approach. The initial step requires manually isolating the region of interest by cropping out the affected tissue area. This cropped image segment is then processed by a specialized tool that converts the pixel data into a corresponding binary matrix (composed of 0s and 1s). The core diagnostic principle hinges on analyzing the statistical distribution of these binary values within the matrix. Specifically, a high frequency or dense concentration of transitions between 0 and 1 (interpreted as a "high number of 01") is proposed to indicate more complex, heterogeneous structures typical of malignant pathologies such as tumors, masses, or cancer. Conversely, a lower frequency of these binary transitions suggests a more uniform internal architecture, which would be characteristic of benign entities like lipomas or simple cysts. Thus, the method aims to differentiate pathological conditions based on quantifiable patterns within the binary representation of the image data.</span><strong><span> </span></strong></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_17862300 |
| institution | Zenodo |
| language | eng |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | BINARY DISTRIBUTION MODEL FOR IDENTIFICATION OF PATHOLOGY IN RADIOLOGICAL IMAGES Umair Masood Awan BINARY DISTRIBUTION MODEL <p><span>Identifying pathology in radiological images involves a mathematical and computational approach. The initial step requires manually isolating the region of interest by cropping out the affected tissue area. This cropped image segment is then processed by a specialized tool that converts the pixel data into a corresponding binary matrix (composed of 0s and 1s). The core diagnostic principle hinges on analyzing the statistical distribution of these binary values within the matrix. Specifically, a high frequency or dense concentration of transitions between 0 and 1 (interpreted as a "high number of 01") is proposed to indicate more complex, heterogeneous structures typical of malignant pathologies such as tumors, masses, or cancer. Conversely, a lower frequency of these binary transitions suggests a more uniform internal architecture, which would be characteristic of benign entities like lipomas or simple cysts. Thus, the method aims to differentiate pathological conditions based on quantifiable patterns within the binary representation of the image data.</span><strong><span> </span></strong></p> |
| title | BINARY DISTRIBUTION MODEL FOR IDENTIFICATION OF PATHOLOGY IN RADIOLOGICAL IMAGES |
| topic | BINARY DISTRIBUTION MODEL |
| url | https://doi.org/10.5281/zenodo.17862300 |