multi-label-classification-on-chexpert

Multi-Label Classification

Results over time

Click legend items to toggle metrics. Hover points for model names.

Leaderboard

Paper	Code	AVERAGE AUC ON 14 LABEL	NUM RADS BELOW CURVE	ModelName	ReleaseDate
Category-Wise Fine-Tuning for Image Multi-label Classification with Partial Labels	✓ Link	0.933		CFT (ensemble) Macao Polytechnic University	2023-11-27
Interpreting chest X-rays via CNNs that exploit hierarchical disease dependencies and uncertainty labels	✓ Link	0.930	2.600	Hierarchical-Learning-V1 (ensemble)	2019-11-15
Large-scale Robust Deep AUC Maximization: A New Surrogate Loss and Empirical Studies on Medical Image Classification	✓ Link	0.930	2.800	DeepAUC-v1	2020-12-06
		0.929	2.800	YWW(ensemble)
		0.929	2.600	Conditional-Training-LSR
Interpreting chest X-rays via CNNs that exploit hierarchical disease dependencies and uncertainty labels	✓ Link	0.929	2.600	Hierarchical-Learning-V4 (ensemble)	2019-11-15
		0.929	2.600	Conditional-Training-LSR-V1
		0.929	2.600	Hierarchical-Learning-V0 (ensemble)
		0.928	2.600	Multi-Stage-Learning-CNN-V3 (ensemble)
		0.928	2.600	DeepCNNsGM(ensemble)
		0.927	3.000	inisis
		0.927	2.600	DeepCNNs(ensemble)
		0.927	2.600	SenseXDR
		0.927	2.600	ihil (ensemble)
		0.926	3.000	JF aboy ensemble_V2 JF HEALTHCARE https://github.com/deadpoppy/CheXpert-Challeng
		0.926	2.600	DRNet (ensemble)
		0.926	2.600	yw
Anatomy-XNet: An Anatomy Aware Convolutional Neural Network for Thoracic Disease Classification in Chest X-rays		0.926	2.600	Anatomy-XNet-V1	2021-06-10
		0.925	2.400	hoanganh_VB_ensemble3
		0.925	2.400	alimebkovk
		0.924	2.600	uest
		0.924	2.400	Hoang_VB_ensemble31_v
		0.924	2.400	tedtt
		0.924	2.400	as-hust-v3
		0.924	2.400	hoanganh_VB_VN
		0.924	2.400	Hierarchical-CNN-Ensemble-V1 (ensemble)
		0.923	2.600	DE_APR ensemble ltt
		0.923	2.600	DE_APR_N ensemble ltt
		0.923	2.600	Multi-Stage-Learning-CNN-V2 (ensemble)
		0.923	2.600	Weighted-CNN(ensemble)
		0.923	2.400	hoanganhcnu_ensemble27_v
		0.923	2.400	YJ&&YWW :https://github.com/inisis/chexper
		0.923	2.400	as-hust-v1
		0.923	2.400	Maxium (ensemble)
		0.922	2.800	as-hust-v2
		0.922	2.400	Average-CNN(ensemble)
		0.922	2.400	MaxAUC
		0.921	2.600	zjr(ensembel)
		0.921	2.400	SuperCNNv3
		0.921	2.400	hyc
		0.921	2.400	hoangnguyenkcv1
		0.920	2.600	{"submit_id": "0x3c7b0af1b5784c159daf259c58543aa3", "predict_id": "0x67b23473183f4f43afa3b37edbc5d7fe", "submitter_id": "0x30db016ad564455ba055eb7f7f4402ac"
		0.920	2.400	HOANG_VB_VN_2
		0.919	2.600	BDNB
		0.919	2.600	JF Coolver ensemble
		0.919	2.400	thang ensemble colo
		0.919	2.400	hoangnn9 ensemble VBV
		0.919	2.400	JF aboy ensemble_V1 JF HEALTHCARE https://github.com/deadpoppy/CheXpert-Challeng
		0.919	2.200	{"submit_id": "0x33aeb0f2525e482a886196c273bdf1ba", "predict_id": "0xff2f60907da8440d98ff17f0af749535", "submitter_id": "0x0b382a226d4548c9b441f19b1907fe0f"
		0.919	2.200	brian-baseline-v2
		0.918	2.600	DE_JUN4_RS_EN ensemble LTT
		0.918	2.600	Mehdi_You (ensemble)
Category-Wise Fine-Tuning for Image Multi-label Classification with Partial Labels	✓ Link	0.918	2.600	A Good Model (single model) Macao Polytechnic University	2023-11-27
		0.918	2.600	A Good Model (single model)
Anatomy-XNet: An Anatomy Aware Convolutional Neural Network for Thoracic Disease Classification in Chest X-rays		0.917	2.600	Anatomy-XNet (ensemble)	2021-06-10
		0.917	2.400	Ensemble_v2
		0.917	2.200	Deep-CNNs-V1
		0.917	2.200	vdn6 ensemble ltt
		0.917	2.200	Overfit ensemble OTH-A
		0.917	2.000	thangbk(ensemble)
		0.916	2.600	desmond
		0.916	2.600	DE_JUN1_RS_EN ensemble LTT
		0.916	2.400	DE_JUN3_RS_EN ensemble LTT
		0.916	2.400	{"submit_id": "0x57dc2989f0474ca095d0841df09cfb18", "predict_id": "0xd43bcf7d4c9b467894db2b274b18794e", "submitter_id": "0x30db016ad564455ba055eb7f7f4402ac"
		0.916	2.400	ATT-AW-v1
		0.916	2.200	{"submit_id": "0xeb9c9e79ed9e4410a2a37d62322f4585", "predict_id": "0x735b718280b14e83895decbc31641f87", "submitter_id": "0x30db016ad564455ba055eb7f7f4402ac"
		0.916	2.200	Multi-Stage-Learning-CNN-V0
		0.915	2.600	TGNB
		0.915	2.400	ensemble SN
		0.915	2.400	zhangjingyan
		0.915	2.200	Deadpoppy Ensemble
		0.915	2.200	hoangnguyenkcv-ensemble28
		0.914	2.600	DE_JUN2_RS_EN ensemble LTT
		0.914	2.400	GRNB
		0.914	2.000	Deep-CNNs (ensemble)
		0.913	2.200	Sky-Model
		0.913	2.200	JF Deadpoppy
		0.913	2.000	YWW-YJ:https://github.com/inisis/chexper
		0.912	2.200	zjy
		0.912	2.000	WL_Baseline (ensemble)
		0.911	2.200	KCV-CNN-ensemble-CN
		0.911	2.200	songta
		0.911	2.200	bhtrun
		0.911	2.200	anatomy_xnet_v1 (single model)
		0.911	2.000	DS_APR_N single model ltt
		0.911	2.000	DS_APR single model LTT
		0.911	2.000	brian-baseline
		0.910	2.200	ensemble SNU
		0.909	2.200	HinaNetV2 (ensemble)
		0.909	2.000	KD-Prune10 (Single model)
		0.909	1.800	G_Mans_ensembl
Masks and Manuscripts: Advancing Medical Pre-training with End-to-End Masking and Narrative Structuring		0.909		Masks and Manuscripts	2024-07-23
		0.908	2.000	guran_ri
		0.908	1.800	vdnnn (ensemble)
		0.908	1.800	BAAZT
CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison	✓ Link	0.907	1.800	Stanford Baseline (ensemble)	2019-01-21
		0.907	1.600	vbn (single model)
		0.907	1.600	muti_base (ensemble)
		0.907	1.400	Z_Ensemble_V
		0.906	1.600	{ForwardModelEnsembleCorrected} (ensemble)
		0.906	1.600	LBC-v2 (ensemble)
		0.906	1.600	LBC-v2
Image Projective Transformation Rectification with Synthetic Data for Smartphone-captured Chest X-ray Photos Classification	✓ Link	0.906	1.600	LBC-v2 (ensemble)	2022-10-12
		0.905	2.000	Multi-CNN
		0.905	1.800	hy
		0.905	1.600	ForwardMECorrectedFull (ensemble)
		0.904	1.200	JustAnotherDensenet
		0.903	1.600	Orlando (single model)
		0.902	2.000	Max (single model)
		0.902	1.800	DeepLungsEnsemble
		0.901	1.600	Ensemble_v1
		0.901	1.400	Nakajima_ayas
		0.900	1.600	MLC11 NotDense (single-model)
		0.900	1.200	vn_2 single_model ltt
		0.899	2.000	{AVG_MAX}(ensemble)
		0.899	1.800	Z_Ensemble_
		0.899	1.600	llllldz
		0.899	1.600	DiseaseNet Samg2003 single model, UIUC, http://sambhavgupta.com
		0.899	1.600	DiseaseNet Samg2003 single model, DPS RKP, http://sambhavgupta.co
		0.899	1.400	LBC-v0
		0.899	1.400	LBC-v0 (ensemble)
Image Projective Transformation Rectification with Synthetic Data for Smartphone-captured Chest X-ray Photos Classification	✓ Link	0.899	1.400	LBC-v0 (ensemble)	2022-10-12
		0.898	1.800	BUA
		0.898	1.400	G_Mans_v2 (single model): LibAUC + coat_mini
		0.898	1.200	ljc226
		0.897	1.600	ForwardModelEnsemble (ensemble)
		0.897	1.600	NewTrickTest (ensemble)
		0.897	1.200	AccidentNet v1 (single model)
		0.896	1.600	ylz-v01
		0.896	1.400	ldz
		0.896	1.400	Densenet
		0.896	1.400	Stellarium-CheXpert-Local (single model)
		0.896	1.400	Stellarium-CheXpert-Local
Image Projective Transformation Rectification with Synthetic Data for Smartphone-captured Chest X-ray Photos Classification	✓ Link	0.896	1.400	Stellarium-CheXpert-Local	2022-10-12
		0.895	1.800	Deadpoppy Single
		0.895	1.600	adoudo
		0.895	1.400	{koala-large} (single model)
		0.895	1.200	MVD121
		0.895	1.000	hust(single model)
		0.894	1.600	MM1
		0.894	1.600	hycN
		0.894	1.600	zhujier
		0.894	1.000	U-Random-Ind (single)
		0.892	1.600	HybridModelEnsemble (ensemble)
		0.891	1.200	MVD121-320
		0.891	1.000	ylz-v02
		0.890	1.000	pause
		0.890	1.000	Overfit ensemble OT
		0.890	0.800	Haruka_Hamasak
		0.889	1.400	DenseNet169 at 320x320 (single model)
		0.889	1.400	LR-baseline (ensemble)
		0.888	1.000	DataAugFTW (single model)
		0.888	1.000	{koala} (single model)
		0.887	1.200	Xception (single model)
		0.887	1.200	Stellarium (single model)
		0.887	1.200	Stellarium
		0.887	1.200	pm_rn50_0.15pp
		0.886	1.200	baseline3
		0.886	1.000	PrateekMunja
		0.886	0.800	MVR50
		0.884	1.600	MNet-Fix (Single Model)
		0.884	0.800	Coolver XH
		0.883	1.200	Naive Densenet
		0.883	0.600	mhealth_buet (single model)
		0.882	0.800	Aoitori (single model)
		0.882	0.600	{chexpert-classifier}(single model)
		0.882	0.400	DearBrave (single model)
		0.881	1.000	AccidentNet V2 (single model)
		0.880	1.200	{densenet} (single model)
		0.879	0.600	Yoake (single model)
		0.878	0.600	MLC11 Baseline (single-model)
		0.876	1.200	DenseNet
		0.876	1.000	HCL1 (single model)
		0.875	1.200	MLGCN (single model)
		0.875	1.000	GCN_densenet121-single mode
		0.873	0.800	GreenTeaCalpis (single model)
		0.873	0.400	Multi-CNN (ensemble)
		0.871	0.600	BASELINE ResNet50
		0.868	0.800	baseline1 (single model)
		0.868	0.600	Baseline DenseNet161
		0.865	0.600	DSENet
		0.863	0.800	Densenet-Basic Single NUS
		0.862	0.800	KD_Mobilenet (single model)
		0.861	1.000	{GoDense} (single model)
		0.861	0.400	inceptionv3_single_NN
		0.860	0.800	MLKD (Single model)
		0.860	0.600	BASELINE Acorn
		0.859	0.600	ErrorNet (single model)
		0.859	0.600	SleepNet (single model)
		0.858	1.000	baseline2
		0.858	0.000	UMLS_CLIP (single model)
		0.854	0.800	haw02 (single model)
		0.853	0.000	CombinedTrainDenseNet121 (single model)
		0.851	0.400	rayOfLightSingle (Single Model)
		0.850	0.600	Model_Team_34 (single model)
		0.850	0.400	Test model habbe
		0.848	0.600	model2_DenseNet121
		0.848	0.200	Baseline
		0.844	0.400	HinaNet (single model)
		0.842	0.200	singlehead_models (single model combined)
		0.840	0.400	mwowra-conditional (single)
		0.838	0.400	multihead_model (one model for all pathologies)
		0.837	0.200	mobilenet (single model)
		0.835	0.000	Grp12BigCNN
		0.834	0.400	MLC9_Densenet (single model)
		0.830	0.200	Grp12v2USup2OSamp (ensemble)
		0.822	0.000	DNET121-single
CheXclusion: Fairness gaps in deep chest X-ray classifiers	✓ Link	0.805		DensNet121	2020-02-14
		0.797	0.600	G_Mans_v1 (single model):
		0.769	0.000	12ASLv2(single)
		0.760	0.000	DenseNet121 (single model)
		0.736	0.000	12ASLv1(single)
		0.732	0.600	haw-baseline (single model)
		0.727	0.000	rayOfLight (ensemble)
		0.724	0.000	BASELINE DenseNet121
		0.618	0.200	Chest-x-ray classification using
		0.615	0.000	BME_Final_v2
		0.606	0.000	{densenet121}{single model
		0.606	0.000	autobot
		0.575	0.000	{MLC02_DenseNet121}
		0.531	0.000	efficiantB5 (single model)
		0.524	0.000	apalepu1
		0.500	0.000	Erdem (single)
		0.481	0.000	Adalab Standard (Single Model)
		0.481	0.000	Adalab Standard (single model)
		0.479	0.000	zeroshot_medclip_baseline (ensemble)

OpenCodePapers

multi-label-classification-on-chexpert