Medical image classifiers typically return predictions without a native contract for independently reconstructing their spatial decision computation. Evidence Field is a trainable spatial-program architecture that jointly produces class scores, selected covering programs and signed evidence. Its encoder combines absolute and image-relative response fields with associative spatial moments and learned characteristic sums. A typed classifier scores individual regions and parent-conditioned multiplicative groups; a covering recursion selects scale, partition and composition. We establish exact one-fold coverage, optimality within the registered grammar, a bound relating smooth training to hard inference, signed-score conservation and class-centred offset invariance. Direct contractions and their adjoint specify arithmetic and storage scaling on two- and three-dimensional grids. Native output records bind inputs, parameters, supports, types and contextual dependencies to individual and ensemble predictions. In a separately reported histopathology application, patient-separated training and evaluation across held-out centres provide empirical evidence for the architecture. Trained mechanism and corrected-loss comparisons distinguish predictive effects from implementation correctness; independent readers reconstruct descriptors, routes, scores and probability mixtures. The principal contribution is a coupled representation, decision and evidence architecture whose selected computation can be reproduced independently.
Figures & tables
Fig. 1: EF architecture, read bottom to top. Ck(v) is the best partition/packing continuation score; tracing Fk yields program Πk . The right branch exposes training and independent replay.
Symbols
Meaning
d,N
Grid dimension and number of samples in Ω .
R,L,K
Response fields, learned frequencies and output classes.
H,A,b,J
Grammar depth, partition alternatives, b=2d children, maximum group order.
Mn,Mg,D
Own/group score types and region descriptor width.
v,π(v),mv,ℓv
Region occurrence, parent, midpoint and half-width vector.
TABLE I: Notation used by the architecture and its equations.
Figure 3
Fig. 4: Complete-score operands, partial composition occupancies, signed conservation and class-centred alignment. Group scores retain their member anchors.
Evaluation
N (patients)
EF
CNN
ViT
CAMELYON centre 1
34,904 (8)
0.9356 / 84.46
0.7352 / 58.94
0.8502 / 76.13
CAMELYON centre 2
85,054 (9)
0.9453 / 86.77
0.7656 / 67.60
0.7924 / 67.27
TABLE II: Histopathology companion: raw three-member ensemble AUROC / accuracy (%), fixed argmax. N counts patches, parentheses patients; EF is source-selected paired CE.
TABLE IV: Hash-bound CAMELYON17 implementation audits: two source loss inputs, both held-out centres for replay. Finite-difference error is relative; others absolute. Loss checks have no optimizer steps.
Fig. 5: Worked evidence: central- 322y=0 , seed 11’s 962 class-1 cover and signed margin N(E10−E00) . Units are supports, not cells. Raw ensemble Q1=0.88756339 is incorrect.
Deep neural networks can achieve strong predictive performance in medical imaging while relying on regions unrelated to the predicted finding. Explanation-guided learning can address this problem, but multi-label classification introduces a key challenge: multiple abnormalities may coexist in an image, requiring each prediction to align with its corresponding spatial evidence. We propose a class-conditioned explanation-guided learning framework that constructs a separate explanation for each positive class and penalizes evidence outside its class-specific annotated regions. We instantiate the framework using Gradient-weighted Class Activation Mapping (Grad-CAM) and evaluate it on 14 abnormalities using EfficientNet-B1, ViT-B/16, ViT-B/32, and Swin-B. Explanation guidance substantially improves annotation alignment across all architectures: All-Saliency Precision increases from 0.140 to 0.625, 0.052 to 0.551, 0.060 to 0.403, and 0.095 to 0.570, respectively. These improvements are accompanied by smaller, architecture-dependent changes in classification performance. Our results show that class-conditioned explanation guidance can substantially improve where multi-label classifiers concentrate their explanatory evidence while largely preserving predictive performance.
Zubair Faruqui, Rahul Dubey
Department of Computer Science, Missouri State University
Segment-based explanation methods such as LIME return a single explanation for each prediction, computed from one fixed image segmentation. This hides two important facts: a prediction can be supported by many different sets of image segments, and the segmentation itself shapes which explanations can be found. We introduce MOXIE (Multi-Objective eXplanation Imaging Engine), an evolutionary framework that searches for segment subsets that preserve the classifier's confidence while keeping as little of the image as possible. Instead of one explanation, MOXIE returns a Pareto front of alternative explanations that range from compact to highly faithful. We evaluate MOXIE with NSGA-II and four segmentation methods (SLIC, Felzenszwalb, Watershed and Voronoi) on BloodMNIST and HAM10000 datasets, using the same evaluation budget as LIME. Results show that MOXIE achieves a higher hypervolume than LIME on every image. LIME's explanations often appear convincing, yet the classifier's confidence collapses when only the highlighted segments are shown. MOXIE's fronts reveal how much of the image is needed to preserve the model's confidence and which contextual regions influence it. We also find that segmentation strongly affects evaluation: methods with unequal segment sizes appear most compact when segments are counted. These results show that alternative explanations provide a more complete view of a model's decision than a single explanation.
Abiha Tahsin Chowdhury, Rahul Dubey
Department of Computer Science, Missouri State University, Springfield, MO, USA
Deep learning models for medical imaging often exhibit overconfidence, creating safety risks in ambiguous diagnostic scenarios. While Conformal Prediction (CP) provides distribution-free statistical guarantees, standard methods such as Regularized Adaptive Prediction Sets (RAPS) optimize for average efficiency and can mask severe failures on difficult inputs. We propose an Adaptive Lambda Criterion for RAPS that minimizes the worst-case coverage violation across prediction set size strata. On OrganAMNIST (58,850 abdominal CT images, 11 classes), standard size-optimized RAPS converges to near-deterministic behavior with stratified undercoverage on uncertain samples, while our method achieves 95.72 percent global coverage with average set size 1.09 and at least 90 percent coverage across all strata. Cross-domain validation on PathMNIST (107,180 pathology images, 9 classes) confirms generalizability. Quantitative Grad-CAM analysis (rho = -0.30, p < 1e-22) shows that multi-label predictions correspond to focused attention on anatomically ambiguous regions. These results demonstrate that the proposed method improves reliability while maintaining efficiency, making it suitable for safety-critical medical AI applications.
One Octadion, Novanto Yudistira, Lailil Muflikhah
Faculty of Computer Science, Universitas Brawijaya, Malang, Indonesia