Data-Aware Layer Assignment for Secure and EfficientCommunication in Federated Learning for Medical ImageAnalysis Article Swipe
Sai Sriram Gonthina
,
Sandip Roy
,
Pronaya Bhattacharya
,
Pratip Rana
,
Sachin Shetty
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.1609/aaaiss.v7i1.36926
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.1609/aaaiss.v7i1.36926
Cross-silo medical imaging federations must contend with strict privacy, limited bandwidth, and non identically distributed (non-IID) data that destabilize training. Current federated learning (FL) architectures either carry the full model (e.g., FedAvg/FedProx) or use naive client/layer pruning and random sampling while ignoring both non-IID heterogeneity and per-layer utility. Based on these limitations, the paper presents a dataaware, layer-wise protocol that aligns communication with expected loss descent while bounding per-round client leverage. Each round, the server estimates per-layer influence from a tiny root set, and clients expose lightweight metadata to form data-quality scores. A capacity-constrained entropic transport matches high-influence layers to high-quality clients under redundancy and temporal coverage. Clients train all layers but upload exactly one with train-all, send-one principle. The server then performs per-layer robust aggregation on masked updates via secure aggregation. On the three cross-silo imaging benchmarks of Pneumonia CXR, Brain-Tumor MRI, and ISIC Skin Cancer, it demonstrates a strong threshold free detection quality (AUROC/ AUPRC: 0.925/0.935, 0.996/0.988, 0.834/0.852, respectively) while also reducing the per round up-link by ≈ 1/n with respect to FedAvg (e.g., ≈ 10× with 10 clients) by only receiving one layer per client. Indicating its viability for deployment-grade secure aggregation for hospital networks.
Related Topics
Concepts
No concepts available.
Metadata
- Type
- article
- Landing Page
- https://doi.org/10.1609/aaaiss.v7i1.36926
- https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064
- OA Status
- diamond
- OpenAlex ID
- https://openalex.org/W4416553130
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4416553130Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.1609/aaaiss.v7i1.36926Digital Object Identifier
- Title
-
Data-Aware Layer Assignment for Secure and EfficientCommunication in Federated Learning for Medical ImageAnalysisWork title
- Type
-
articleOpenAlex work type
- Publication year
-
2025Year of publication
- Publication date
-
2025-11-23Full publication date if available
- Authors
-
Sai Sriram Gonthina, Sandip Roy, Pronaya Bhattacharya, Pratip Rana, Sachin ShettyList of authors in order
- Landing page
-
https://doi.org/10.1609/aaaiss.v7i1.36926Publisher landing page
- PDF URL
-
https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
diamondOpen access status per OpenAlex
- OA URL
-
https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4416553130 |
|---|---|
| doi | https://doi.org/10.1609/aaaiss.v7i1.36926 |
| ids.doi | https://doi.org/10.1609/aaaiss.v7i1.36926 |
| ids.openalex | https://openalex.org/W4416553130 |
| fwci | |
| type | article |
| title | Data-Aware Layer Assignment for Secure and EfficientCommunication in Federated Learning for Medical ImageAnalysis |
| biblio.issue | 1 |
| biblio.volume | 7 |
| biblio.last_page | 523 |
| biblio.first_page | 516 |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | |
| locations[0].id | doi:10.1609/aaaiss.v7i1.36926 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4389157828 |
| locations[0].source.issn | 2994-4317 |
| locations[0].source.type | journal |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | 2994-4317 |
| locations[0].source.is_core | True |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | Proceedings of the AAAI Symposium Series |
| locations[0].source.host_organization | |
| locations[0].source.host_organization_name | |
| locations[0].license | |
| locations[0].pdf_url | https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064 |
| locations[0].version | publishedVersion |
| locations[0].raw_type | journal-article |
| locations[0].license_id | |
| locations[0].is_accepted | True |
| locations[0].is_published | True |
| locations[0].raw_source_name | Proceedings of the AAAI Symposium Series |
| locations[0].landing_page_url | https://doi.org/10.1609/aaaiss.v7i1.36926 |
| indexed_in | crossref |
| authorships[0].author.id | https://openalex.org/A5119273966 |
| authorships[0].author.orcid | |
| authorships[0].author.display_name | Sai Sriram Gonthina |
| authorships[0].countries | IN |
| authorships[0].affiliations[0].institution_ids | https://openalex.org/I4210138251, https://openalex.org/I4387155265 |
| authorships[0].affiliations[0].raw_affiliation_string | IIIT Naya Raipur |
| authorships[0].institutions[0].id | https://openalex.org/I4387155265 |
| authorships[0].institutions[0].ror | https://ror.org/016kfyg29 |
| authorships[0].institutions[0].type | facility |
| authorships[0].institutions[0].lineage | https://openalex.org/I4387155265 |
| authorships[0].institutions[0].country_code | |
| authorships[0].institutions[0].display_name | International Institute of Information Technology |
| authorships[0].institutions[1].id | https://openalex.org/I4210138251 |
| authorships[0].institutions[1].ror | https://ror.org/03e2e3s57 |
| authorships[0].institutions[1].type | education |
| authorships[0].institutions[1].lineage | https://openalex.org/I4210138251 |
| authorships[0].institutions[1].country_code | IN |
| authorships[0].institutions[1].display_name | Indian Institute of Information Technology, Nagpur |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Sai Sriram Gonthina |
| authorships[0].is_corresponding | False |
| authorships[0].raw_affiliation_strings | IIIT Naya Raipur |
| authorships[1].author.id | https://openalex.org/A5083317923 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-5447-803X |
| authorships[1].author.display_name | Sandip Roy |
| authorships[1].countries | GH, US |
| authorships[1].affiliations[0].institution_ids | https://openalex.org/I4210092569, https://openalex.org/I81365321 |
| authorships[1].affiliations[0].raw_affiliation_string | Old Dominion University |
| authorships[1].institutions[0].id | https://openalex.org/I4210092569 |
| authorships[1].institutions[0].ror | https://ror.org/003kqe171 |
| authorships[1].institutions[0].type | education |
| authorships[1].institutions[0].lineage | https://openalex.org/I4210092569 |
| authorships[1].institutions[0].country_code | GH |
| authorships[1].institutions[0].display_name | Dominion University College |
| authorships[1].institutions[1].id | https://openalex.org/I81365321 |
| authorships[1].institutions[1].ror | https://ror.org/04zjtrb98 |
| authorships[1].institutions[1].type | education |
| authorships[1].institutions[1].lineage | https://openalex.org/I81365321 |
| authorships[1].institutions[1].country_code | US |
| authorships[1].institutions[1].display_name | Old Dominion University |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Sandip Roy |
| authorships[1].is_corresponding | False |
| authorships[1].raw_affiliation_strings | Old Dominion University |
| authorships[2].author.id | https://openalex.org/A5053527968 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-1206-2298 |
| authorships[2].author.display_name | Pronaya Bhattacharya |
| authorships[2].affiliations[0].raw_affiliation_string | Amity University |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Pronaya Bhattacharya |
| authorships[2].is_corresponding | False |
| authorships[2].raw_affiliation_strings | Amity University |
| authorships[3].author.id | https://openalex.org/A5018203536 |
| authorships[3].author.orcid | https://orcid.org/0000-0001-9199-2479 |
| authorships[3].author.display_name | Pratip Rana |
| authorships[3].countries | GH, US |
| authorships[3].affiliations[0].institution_ids | https://openalex.org/I4210092569, https://openalex.org/I81365321 |
| authorships[3].affiliations[0].raw_affiliation_string | Old Dominion University |
| authorships[3].institutions[0].id | https://openalex.org/I4210092569 |
| authorships[3].institutions[0].ror | https://ror.org/003kqe171 |
| authorships[3].institutions[0].type | education |
| authorships[3].institutions[0].lineage | https://openalex.org/I4210092569 |
| authorships[3].institutions[0].country_code | GH |
| authorships[3].institutions[0].display_name | Dominion University College |
| authorships[3].institutions[1].id | https://openalex.org/I81365321 |
| authorships[3].institutions[1].ror | https://ror.org/04zjtrb98 |
| authorships[3].institutions[1].type | education |
| authorships[3].institutions[1].lineage | https://openalex.org/I81365321 |
| authorships[3].institutions[1].country_code | US |
| authorships[3].institutions[1].display_name | Old Dominion University |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Pratip Rana |
| authorships[3].is_corresponding | False |
| authorships[3].raw_affiliation_strings | Old Dominion University |
| authorships[4].author.id | https://openalex.org/A5052787847 |
| authorships[4].author.orcid | https://orcid.org/0000-0002-8789-0610 |
| authorships[4].author.display_name | Sachin Shetty |
| authorships[4].countries | GH, US |
| authorships[4].affiliations[0].institution_ids | https://openalex.org/I4210092569, https://openalex.org/I81365321 |
| authorships[4].affiliations[0].raw_affiliation_string | Old Dominion University |
| authorships[4].institutions[0].id | https://openalex.org/I4210092569 |
| authorships[4].institutions[0].ror | https://ror.org/003kqe171 |
| authorships[4].institutions[0].type | education |
| authorships[4].institutions[0].lineage | https://openalex.org/I4210092569 |
| authorships[4].institutions[0].country_code | GH |
| authorships[4].institutions[0].display_name | Dominion University College |
| authorships[4].institutions[1].id | https://openalex.org/I81365321 |
| authorships[4].institutions[1].ror | https://ror.org/04zjtrb98 |
| authorships[4].institutions[1].type | education |
| authorships[4].institutions[1].lineage | https://openalex.org/I81365321 |
| authorships[4].institutions[1].country_code | US |
| authorships[4].institutions[1].display_name | Old Dominion University |
| authorships[4].author_position | last |
| authorships[4].raw_author_name | Sachin Shetty |
| authorships[4].is_corresponding | False |
| authorships[4].raw_affiliation_strings | Old Dominion University |
| has_content.pdf | True |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064 |
| open_access.oa_status | diamond |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-11-23T00:00:00 |
| display_name | Data-Aware Layer Assignment for Secure and EfficientCommunication in Federated Learning for Medical ImageAnalysis |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-28T17:12:10.076149 |
| primary_topic | |
| cited_by_count | 0 |
| locations_count | 1 |
| best_oa_location.id | doi:10.1609/aaaiss.v7i1.36926 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4389157828 |
| best_oa_location.source.issn | 2994-4317 |
| best_oa_location.source.type | journal |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | 2994-4317 |
| best_oa_location.source.is_core | True |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | Proceedings of the AAAI Symposium Series |
| best_oa_location.source.host_organization | |
| best_oa_location.source.host_organization_name | |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064 |
| best_oa_location.version | publishedVersion |
| best_oa_location.raw_type | journal-article |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | True |
| best_oa_location.is_published | True |
| best_oa_location.raw_source_name | Proceedings of the AAAI Symposium Series |
| best_oa_location.landing_page_url | https://doi.org/10.1609/aaaiss.v7i1.36926 |
| primary_location.id | doi:10.1609/aaaiss.v7i1.36926 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4389157828 |
| primary_location.source.issn | 2994-4317 |
| primary_location.source.type | journal |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | 2994-4317 |
| primary_location.source.is_core | True |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | Proceedings of the AAAI Symposium Series |
| primary_location.source.host_organization | |
| primary_location.source.host_organization_name | |
| primary_location.license | |
| primary_location.pdf_url | https://ojs.aaai.org/index.php/AAAI-SS/article/download/36926/39064 |
| primary_location.version | publishedVersion |
| primary_location.raw_type | journal-article |
| primary_location.license_id | |
| primary_location.is_accepted | True |
| primary_location.is_published | True |
| primary_location.raw_source_name | Proceedings of the AAAI Symposium Series |
| primary_location.landing_page_url | https://doi.org/10.1609/aaaiss.v7i1.36926 |
| publication_date | 2025-11-23 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.A | 92 |
| abstract_inverted_index.a | 55, 79, 149 |
| abstract_inverted_index.10 | 179 |
| abstract_inverted_index.On | 132 |
| abstract_inverted_index.by | 168, 181 |
| abstract_inverted_index.it | 147 |
| abstract_inverted_index.of | 138 |
| abstract_inverted_index.on | 49, 126 |
| abstract_inverted_index.or | 32 |
| abstract_inverted_index.to | 88, 99, 173 |
| abstract_inverted_index.1/n | 170 |
| abstract_inverted_index.The | 119 |
| abstract_inverted_index.all | 109 |
| abstract_inverted_index.and | 11, 37, 45, 83, 104, 143 |
| abstract_inverted_index.but | 111 |
| abstract_inverted_index.for | 191, 195 |
| abstract_inverted_index.its | 189 |
| abstract_inverted_index.non | 12 |
| abstract_inverted_index.one | 114, 184 |
| abstract_inverted_index.per | 165, 186 |
| abstract_inverted_index.the | 27, 52, 73, 133, 164 |
| abstract_inverted_index.use | 33 |
| abstract_inverted_index.via | 129 |
| abstract_inverted_index.≈ | 169, 176 |
| abstract_inverted_index.(FL) | 23 |
| abstract_inverted_index.10× | 177 |
| abstract_inverted_index.CXR, | 140 |
| abstract_inverted_index.Each | 71 |
| abstract_inverted_index.ISIC | 144 |
| abstract_inverted_index.MRI, | 142 |
| abstract_inverted_index.Skin | 145 |
| abstract_inverted_index.also | 162 |
| abstract_inverted_index.both | 42 |
| abstract_inverted_index.data | 16 |
| abstract_inverted_index.form | 89 |
| abstract_inverted_index.free | 152 |
| abstract_inverted_index.from | 78 |
| abstract_inverted_index.full | 28 |
| abstract_inverted_index.loss | 64 |
| abstract_inverted_index.must | 4 |
| abstract_inverted_index.only | 182 |
| abstract_inverted_index.root | 81 |
| abstract_inverted_index.set, | 82 |
| abstract_inverted_index.that | 17, 59 |
| abstract_inverted_index.then | 121 |
| abstract_inverted_index.tiny | 80 |
| abstract_inverted_index.with | 6, 62, 115, 171, 178 |
| abstract_inverted_index.Based | 48 |
| abstract_inverted_index.carry | 26 |
| abstract_inverted_index.layer | 185 |
| abstract_inverted_index.model | 29 |
| abstract_inverted_index.naive | 34 |
| abstract_inverted_index.paper | 53 |
| abstract_inverted_index.round | 166 |
| abstract_inverted_index.these | 50 |
| abstract_inverted_index.three | 134 |
| abstract_inverted_index.train | 108 |
| abstract_inverted_index.under | 102 |
| abstract_inverted_index.while | 40, 66, 161 |
| abstract_inverted_index.(e.g., | 30, 175 |
| abstract_inverted_index.AUPRC: | 156 |
| abstract_inverted_index.FedAvg | 174 |
| abstract_inverted_index.aligns | 60 |
| abstract_inverted_index.client | 69 |
| abstract_inverted_index.either | 25 |
| abstract_inverted_index.expose | 85 |
| abstract_inverted_index.layers | 98, 110 |
| abstract_inverted_index.masked | 127 |
| abstract_inverted_index.random | 38 |
| abstract_inverted_index.robust | 124 |
| abstract_inverted_index.round, | 72 |
| abstract_inverted_index.secure | 130, 193 |
| abstract_inverted_index.server | 74, 120 |
| abstract_inverted_index.strict | 7 |
| abstract_inverted_index.strong | 150 |
| abstract_inverted_index.upload | 112 |
| abstract_inverted_index.(AUROC/ | 155 |
| abstract_inverted_index.Cancer, | 146 |
| abstract_inverted_index.Clients | 107 |
| abstract_inverted_index.Current | 20 |
| abstract_inverted_index.client. | 187 |
| abstract_inverted_index.clients | 84, 101 |
| abstract_inverted_index.contend | 5 |
| abstract_inverted_index.descent | 65 |
| abstract_inverted_index.exactly | 113 |
| abstract_inverted_index.imaging | 2, 136 |
| abstract_inverted_index.limited | 9 |
| abstract_inverted_index.matches | 96 |
| abstract_inverted_index.medical | 1 |
| abstract_inverted_index.non-IID | 43 |
| abstract_inverted_index.pruning | 36 |
| abstract_inverted_index.quality | 154 |
| abstract_inverted_index.respect | 172 |
| abstract_inverted_index.scores. | 91 |
| abstract_inverted_index.up-link | 167 |
| abstract_inverted_index.updates | 128 |
| abstract_inverted_index.bounding | 67 |
| abstract_inverted_index.clients) | 180 |
| abstract_inverted_index.entropic | 94 |
| abstract_inverted_index.expected | 63 |
| abstract_inverted_index.hospital | 196 |
| abstract_inverted_index.ignoring | 41 |
| abstract_inverted_index.learning | 22 |
| abstract_inverted_index.metadata | 87 |
| abstract_inverted_index.performs | 122 |
| abstract_inverted_index.presents | 54 |
| abstract_inverted_index.privacy, | 8 |
| abstract_inverted_index.protocol | 58 |
| abstract_inverted_index.reducing | 163 |
| abstract_inverted_index.sampling | 39 |
| abstract_inverted_index.send-one | 117 |
| abstract_inverted_index.temporal | 105 |
| abstract_inverted_index.utility. | 47 |
| abstract_inverted_index.(non-IID) | 15 |
| abstract_inverted_index.Pneumonia | 139 |
| abstract_inverted_index.coverage. | 106 |
| abstract_inverted_index.detection | 153 |
| abstract_inverted_index.estimates | 75 |
| abstract_inverted_index.federated | 21 |
| abstract_inverted_index.influence | 77 |
| abstract_inverted_index.leverage. | 70 |
| abstract_inverted_index.networks. | 197 |
| abstract_inverted_index.per-layer | 46, 76, 123 |
| abstract_inverted_index.per-round | 68 |
| abstract_inverted_index.receiving | 183 |
| abstract_inverted_index.threshold | 151 |
| abstract_inverted_index.training. | 19 |
| abstract_inverted_index.transport | 95 |
| abstract_inverted_index.viability | 190 |
| abstract_inverted_index.Cross-silo | 0 |
| abstract_inverted_index.Indicating | 188 |
| abstract_inverted_index.bandwidth, | 10 |
| abstract_inverted_index.benchmarks | 137 |
| abstract_inverted_index.cross-silo | 135 |
| abstract_inverted_index.dataaware, | 56 |
| abstract_inverted_index.layer-wise | 57 |
| abstract_inverted_index.principle. | 118 |
| abstract_inverted_index.redundancy | 103 |
| abstract_inverted_index.train-all, | 116 |
| abstract_inverted_index.Brain-Tumor | 141 |
| abstract_inverted_index.aggregation | 125, 194 |
| abstract_inverted_index.destabilize | 18 |
| abstract_inverted_index.distributed | 14 |
| abstract_inverted_index.federations | 3 |
| abstract_inverted_index.identically | 13 |
| abstract_inverted_index.lightweight | 86 |
| abstract_inverted_index.0.834/0.852, | 159 |
| abstract_inverted_index.0.925/0.935, | 157 |
| abstract_inverted_index.0.996/0.988, | 158 |
| abstract_inverted_index.aggregation. | 131 |
| abstract_inverted_index.client/layer | 35 |
| abstract_inverted_index.data-quality | 90 |
| abstract_inverted_index.demonstrates | 148 |
| abstract_inverted_index.high-quality | 100 |
| abstract_inverted_index.limitations, | 51 |
| abstract_inverted_index.architectures | 24 |
| abstract_inverted_index.communication | 61 |
| abstract_inverted_index.heterogeneity | 44 |
| abstract_inverted_index.respectively) | 160 |
| abstract_inverted_index.high-influence | 97 |
| abstract_inverted_index.FedAvg/FedProx) | 31 |
| abstract_inverted_index.deployment-grade | 192 |
| abstract_inverted_index.capacity-constrained | 93 |
| cited_by_percentile_year | |
| countries_distinct_count | 3 |
| institutions_distinct_count | 5 |
| citation_normalized_percentile |