Left: 365 nm, Middle: 405 nm, Right: Visible light LED. All classifiers were trained on the identical set of RBC instances, using images acquired each at their respective wavelength (multiwavelength dataset, see Table 1). (TIF)</p
(left columns: 365 nm, middle columns: 405 nm, right columns: standard visible LED). Four example ce...
<p>The confusion matrix for taxonomy A was constructed with 10-CV and 17 DEGs. The other confusion m...
In all three plots, log-scaled histograms of classifier confidence scores are displayed in the patte...
Equipping a commercial widefield microscope with 405 nm excitation generates high-contrast label-fre...
These confusion matrices show the result of applying the empirically-optimal threshold values of 61%...
Top: Overall accuracy (left), FPR for rings (center), and recall for rings (right) were selected for...
Classification was evaluated with either two, three, or four categories. For each level of granulari...
Since each classifier distinguishes between the desired class and every “other” class, the confusion...
The arrangement of the categories is the same as Fig 2 from the main text. For this montage, a maxim...
The correctly classified data is reflected along the diagonal regions. The misclassified is reflecte...
(a) RF, (b) GBM, (c) AdaBoost, (d) LR, (e) SVC, (f) SVEC-H, (g) SVEC-S, (h) CNN, (i) LSTM, (j) CNN-L...
<p>The rows of the matrix indicate the actual roughness provided to the participants and the columns...
<p>Confusion matrix and overall performance of the classifier used to determine the sharpness of the...
<p>The matrices show the average locations estimated by the classifier as a function of the actual w...
<p>The matrices show the average locations estimated by the classifier as a function of the actual w...
(left columns: 365 nm, middle columns: 405 nm, right columns: standard visible LED). Four example ce...
<p>The confusion matrix for taxonomy A was constructed with 10-CV and 17 DEGs. The other confusion m...
In all three plots, log-scaled histograms of classifier confidence scores are displayed in the patte...
Equipping a commercial widefield microscope with 405 nm excitation generates high-contrast label-fre...
These confusion matrices show the result of applying the empirically-optimal threshold values of 61%...
Top: Overall accuracy (left), FPR for rings (center), and recall for rings (right) were selected for...
Classification was evaluated with either two, three, or four categories. For each level of granulari...
Since each classifier distinguishes between the desired class and every “other” class, the confusion...
The arrangement of the categories is the same as Fig 2 from the main text. For this montage, a maxim...
The correctly classified data is reflected along the diagonal regions. The misclassified is reflecte...
(a) RF, (b) GBM, (c) AdaBoost, (d) LR, (e) SVC, (f) SVEC-H, (g) SVEC-S, (h) CNN, (i) LSTM, (j) CNN-L...
<p>The rows of the matrix indicate the actual roughness provided to the participants and the columns...
<p>Confusion matrix and overall performance of the classifier used to determine the sharpness of the...
<p>The matrices show the average locations estimated by the classifier as a function of the actual w...
<p>The matrices show the average locations estimated by the classifier as a function of the actual w...
(left columns: 365 nm, middle columns: 405 nm, right columns: standard visible LED). Four example ce...
<p>The confusion matrix for taxonomy A was constructed with 10-CV and 17 DEGs. The other confusion m...
In all three plots, log-scaled histograms of classifier confidence scores are displayed in the patte...