{"id":49301137,"url":"https://github.com/gtkacz/transform-ordering-selection","last_synced_at":"2026-04-26T07:01:51.276Z","repository":{"id":230126740,"uuid":"778560120","full_name":"gtkacz/transform-ordering-selection","owner":"gtkacz","description":"Code for my undergraduate thesis: Quantitative Analysis of the Impact of Image Pre-Processing on the Accuracy of Computer Vision Models Trained to Identify Dermatological Skin Diseases","archived":false,"fork":false,"pushed_at":"2026-04-13T03:04:13.000Z","size":3230266,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-04-13T05:08:36.268Z","etag":null,"topics":["cnn","computer-vision","machine-learning","pre-processing","preprocessing"],"latest_commit_sha":null,"homepage":"https://www.overleaf.com/read/kzmvsyrpncgh#ca0f54","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/gtkacz.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2024-03-28T00:18:24.000Z","updated_at":"2026-04-13T03:04:16.000Z","dependencies_parsed_at":"2025-07-08T12:38:20.257Z","dependency_job_id":"b93a8e09-f3e2-4a81-b870-270295614cff","html_url":"https://github.com/gtkacz/transform-ordering-selection","commit_stats":null,"previous_names":["gtkacz/undergrad_thesis","gtkacz/transform-ordering-selection"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/gtkacz/transform-ordering-selection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gtkacz%2Ftransform-ordering-selection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gtkacz%2Ftransform-ordering-selection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gtkacz%2Ftransform-ordering-selection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gtkacz%2Ftransform-ordering-selection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/gtkacz","download_url":"https://codeload.github.com/gtkacz/transform-ordering-selection/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/gtkacz%2Ftransform-ordering-selection/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32288653,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-26T06:26:00.361Z","status":"ssl_error","status_checked_at":"2026-04-26T06:25:58.791Z","response_time":129,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cnn","computer-vision","machine-learning","pre-processing","preprocessing"],"created_at":"2026-04-26T07:01:50.470Z","updated_at":"2026-04-26T07:01:51.270Z","avatar_url":"https://github.com/gtkacz.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003ccenter\u003e\n    \u003cp align=\"center\"\u003e\n        \u003cimg src=\"https://logodownload.org/wp-content/uploads/2017/09/mackenzie-logo-3.png\" style=\"height: 7ch;\"\u003e\u003cbr\u003e\n        \u003ch1 align=\"center\"\u003eSignificance of Transform Ordering and Selection in CNN Preprocessing for Binary Skin Classification\u003c/h1\u003e\n        \u003ch4 align=\"center\"\u003eGabriel Mitelman Tkacz, Gustavo Scalabrini Sampaio, Leandro Augusto da Silva\u003c/a\u003e\u003c/h4\u003e\n        \u003ch4 align=\"center\"\u003eMackenzie Presbyterian University \u0026mdash; S\u0026atilde;o Paulo, Brazil\u003c/h4\u003e\n    \u003c/p\u003e\n\u003c/center\u003e\n\n\u003chr\u003e\n\n## Abstract\n\nImage preprocessing is a near-universal step in medical image classification pipelines, yet its impact on model accuracy remains insufficiently characterized, particularly with respect to the ordering of multi-transform compositions. This study reports an exhaustive combinatorial evaluation of all 65 pipelines constructible from four fundamental transforms \u0026mdash; histogram equalization, intensity normalization, non-local means denoising, and color space conversion \u0026mdash; applied to a four-block convolutional neural network trained on the binary healthy-versus-diseased dermatoscopic classification task formed by using a dataset with 20,000 balanced images. Every pipeline was trained from scratch under five independent random seeds. Against a near-ceiling baseline of 98.09% test accuracy, which arithmetically bounds the detectable positive-\u0026alpha; regime to at most +1.91 pp, no pipeline of length 2 or greater achieves a positive mean accuracy gain, and only equalization alone produces a seed-robust improvement. Holm-corrected permutation tests establish three ordering regularities: pipeline length correlates negatively with accuracy gain (*r* = \u0026minus;0.44), equalization-first placement outperforms alternative first-position assignments by 1.10 pp (Cohen's *d* = 1.22), and the bookend configuration with equalization first and normalization last outperforms its mirror image by 2.00 pp (Cohen's *d* = 2.38). At pipeline *k* = 3, ordering accounts for 58% of accuracy-gain variance and transform selection for 42%. Confusion-matrix decomposition indicates that preprocessing-induced degradation is dominated by bidirectional feature-space damage (78%) rather than recoverable decision-threshold shift (22%). Within the scope of this single network, single binary task, single dataset, and single transform set, the evidence supports a cautious harm-minimization heuristic \u0026mdash; equalization first, normalization last \u0026mdash; rather than a general claim about ordering dominance in medical image preprocessing.\n\n## Keywords\n\nImage preprocessing \u0026middot; Transform ordering \u0026middot; Preprocessing pipeline design \u0026middot; Convolutional neural networks \u0026middot; Medical image classification \u0026middot; Dermatological diagnosis\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgtkacz%2Ftransform-ordering-selection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgtkacz%2Ftransform-ordering-selection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgtkacz%2Ftransform-ordering-selection/lists"}